US20100063820A1 - Correlating video images of lip movements with audio signals to improve speech recognition - Google Patents

Correlating video images of lip movements with audio signals to improve speech recognition Download PDF

Info

Publication number
US20100063820A1
US20100063820A1 US12/553,663 US55366309A US2010063820A1 US 20100063820 A1 US20100063820 A1 US 20100063820A1 US 55366309 A US55366309 A US 55366309A US 2010063820 A1 US2010063820 A1 US 2010063820A1
Authority
US
United States
Prior art keywords
speech recognition
audio
speech
video
signals
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Abandoned
Application number
US12/553,663
Inventor
Nambi Seshadri
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Avago Technologies International Sales Pte Ltd
Original Assignee
Broadcom Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Broadcom Corp filed Critical Broadcom Corp
Priority to US12/553,663 priority Critical patent/US20100063820A1/en
Publication of US20100063820A1 publication Critical patent/US20100063820A1/en
Assigned to BROADCOM CORPORATION reassignment BROADCOM CORPORATION ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: SESHADRI, NAMBI
Assigned to BANK OF AMERICA, N.A., AS COLLATERAL AGENT reassignment BANK OF AMERICA, N.A., AS COLLATERAL AGENT PATENT SECURITY AGREEMENT Assignors: BROADCOM CORPORATION
Assigned to AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD. reassignment AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: BROADCOM CORPORATION
Assigned to BROADCOM CORPORATION reassignment BROADCOM CORPORATION TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENTS Assignors: BANK OF AMERICA, N.A., AS COLLATERAL AGENT
Abandoned legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/24Speech recognition using non-acoustical features
    • G10L15/25Speech recognition using non-acoustical features using position of the lips, movement of the lips or face analysis

Definitions

  • the present invention relates to a method of and an apparatus for using video signals along with audio signals of speech to provide speech recognition, within an environment where speech recognition is necessary.
  • the present invention relates to a method of and a system for using video images of lip movements with audio input signals to improve speech recognition.
  • the present invention can be implemented in a hand held device, and the invention may include discrete devices or may be implemented on a semiconductor substrate such as a silicon chip.
  • Human speech is made up of numerous different sounds and syllables. Often in many languages, different sounds and/or syllables are combined to form words and/or sentences. The combination of the sounds, syllables, words, and sentences forms the basis for oral communication.
  • human speech is recognizable if the speech is clear and comprehensible to another human's ears.
  • human speech can be recognizable by a machine if the audio waves of the speech is received, and the audio waves are recognizable by an algorithm operating within the machine.
  • audio speech recognition by machines has advanced in sophistication, the accuracy of audio speech recognition has room for improvements.
  • One example of the present invention can be a method of speech recognition.
  • the method can include the steps of receiving audio signals from, a speech source, receiving video signals from the speech source, and processing the audio signals and the video signals.
  • the method can also include the steps of converting the audio signals and the video signals to recognizable information, and implementing a task based on the recognizable information.
  • the present invention can relate to a speech recognition device.
  • the device can have an audio signal receiver configured to receive audio signals from a speech source, a video signal receiver configured to receive video signals from the speech source, and a processing unit configured to process the audio signals and the video signals.
  • the device can have a conversion unit configured to convert the audio signals and the video signals to recognizable information, and an implementation unit configured to implement a task based on the recognizable information.
  • another example of the present invention can provide a system for speech recognition.
  • the system can include a first receiving means for receiving audio signals from a speech source, a second receiving means for receiving video signals from the speech source, and a processing means for processing the audio signals and the video signals.
  • the system can have a converting means for converting the audio signals and the video signals to recognizable information, and an implementing means for implementing a task based on the recognizable information.
  • another example of the present invention can be directed to a method of speech recognition.
  • the method can include the steps of receiving audio signals from a speech source, receiving video signals from the speech source, processing the audio signals, and converting the audio signals into recognizable information.
  • the method can have the step of processing the video signals when a segment of the audio signals can not be converted into the recognizable information.
  • the video signals can coincide with the segment of the audio signals that cannot be converted into the recognizable information.
  • the method also can have the steps of converting the processed video signals into the recognizable information, and implementing a task based on the recognizable information.
  • the present invention can be a speech recognition device.
  • the device can have an audio signal receiver configured to receive audio signals from a speech source, a video signal receiver configured to receive video signals from the speech source, a first processing unit configured to process the audio signals, and a first conversion unit configured to convert the audio signals to recognizable information.
  • the device can also have a second processing unit configured to process the video signals when the audio signals cannot be converted into the recognizable information, wherein the video signals coincide with the segment of the audio signals that cannot be converted into the recognizable information, a second conversion unit configured to convert the video signals processed into the recognizable information, and an implementation unit configured to implement a task based on the recognizable information.
  • the present invention can be drawn to a system for speech recognition.
  • the system can include a first receiving means for receiving audio signals from a speech source, a second receiving means for receiving video signals from the speech source, a first processing means for processing the audio signals, and a first converting means for converting the audio signals into recognizable information.
  • the system can also have a second processing means for processing the video signals when a segment of the audio signals can not be converted into the recognizable information, wherein the video signals coincide with the segment of the audio signals that cannot be converted into the recognizable information, a second converting means for converting the video signals processed into the recognizable information, and an implementing means for implementing a task based on the recognizable information.
  • FIG. 1 illustrates one example of a speech recognition device using audio signals and correlating video signals to improve speech recognition, in accordance with the present invention
  • FIG. 2 illustrates a flow chart illustrating one example of a method of speech recognition using audio signals and correlating video signals, in accordance with the present invention
  • FIG. 3 illustrates a flow chart illustrating another example of a method of speech recognition using audio signals and correlating video signals, in accordance with the present invention
  • FIG. 4 illustrates a flow chart of a method of speech recognition using audio signals and correlating video signals, in accordance with the present invention.
  • FIG. 5 illustrates one example of a hardware configuration for speech recognition using audio signals and correlating video signals, in accordance with the present invention.
  • FIG. 1 illustrates one example of a speech recognition device for improving speech recognition with audio input signals and correlating video input signals according to the present invention.
  • FIG. 1 shows a mobile phone 100 having a display screen 101 and a plurality of actuators 102 .
  • the mobile phone 100 can include a lens 103 for receiving images in video and/or still picture format.
  • the lens 103 can capture or receive video images of the movements of a user's lips 104 a , 104 b made while the user is speaking.
  • a user may desire to contact a business associate via an e-mail message.
  • the user can access the mobile phone 100 and can activate the speech recognition system of the present invention.
  • the mobile phone 100 is placed adjacent to the user's face where the lens 103 is positioned in proximity to the user's mouth so that the image of the user's lips can be captured by the speech recognition system.
  • the user can be alerted to commence speaking into the mobile phone 100 .
  • the user for example, can speak and request to send an e-mail message to Jane Doe in which her e-mail address is pre-programmed into the mobile phone 100 .
  • the name James Doe is similarly pre-programmed in the mobile phone 100 .
  • the speech recognition system processes the audio signal input and can convert all part of the audio signal input into recognizable information with the exception of the name Jane Doe.
  • the audio speech recognition feature of the invention cannot ascertain if the audio signal is referring to Jane Doe or James Doe.
  • the speech recognition system of the present invention can access a section of the video input signal corresponding to the audio signal pertaining to Jane Doe. Based on the detected lip movements of the video signals, the present invention can reduce the uncertainty of the audio speech recognition, and therefore can perform speech recognition with the aid of the video signals, and determine that the request is an e-mail message to Jane Doe rather than James Doe. Thereafter, the present invention can implement one or more actions to carry out the spoken request by the user.
  • the one or more actions can be in the forms of commands such as initiating an e-mail application software, creating an outgoing e-mail window, and inserting text, and sending the e-mail message.
  • FIG. 1 illustrates a mobile phone 100 having a lens 103 , wherein the mobile phone 100 can be configured with the speech recognition system of the present invention
  • the speech recognition system using audio signals and correlating video signals of the invention can be configured on a variety of electronic device, either mobile or stationary.
  • the improved speech recognition system of the invention can be configured on at least but not limited to a laptop computer, a PDA, an audio/video recording device, a home computer, a game console, a remote controller, or other comparable device.
  • FIG. 2 illustrates one example of a method of speech recognition using audio input signal and correlating video input signals, in accordance with the present invention. Specifically, FIG. 2 illustrates one example of a method of speech recognition using audio input signals together with correlating video images of lip movements.
  • the method of the present example can be implemented in hardware, or software, or a combination of both hardware and software.
  • FIG. 2 illustrates one example of a method of speech recognition according to the present invention.
  • a device configured to include a speech recognition system can be activated at step 200 of FIG. 2 .
  • the present invention provides a user with the option to activate the speech recognition feature when necessary.
  • a detecting sensor along with an optical pick-up such as a lens, can detect for video images resembling the user's lips at step 201 . If the detecting sensor and the lens do not detect images of the user's lips, then the speech recognition system can alert the user to readjust the lens or the device, or reposition the user's lips so that an image can be detected, at step 202 of FIG. 2 . If however, the user's lips can be detected or captured by the lens and the sensor, then the speech recognition system can alert the speaker in step 203 that the speech recognition system is in ready mode, and therefore the user can commence speaking.
  • the speech recognition system of the present invention can commence receiving both audio and video input signals from the user's speech and the user's lip movements at step 204 .
  • the speech recognition system can process the audio input signals corresponding to the speech first.
  • both the audio speech and the correlating images of the user's lip movements can be received by the speech recognition system of a device.
  • the speech recognition system can preliminarily initiate only the audio speech recognition portion of the system, and can preliminarily process only the audio portion of the speech.
  • the present invention at step 206 can recognize the speech as comprehensible and recognizable information using only the audio speech recognition portion of the system without the need to activate the assistance of the video signals.
  • the speech recognition system can process the audio input signals and determine if the audio input signals are recognizable as speech at step 205 . If it is determined that the audio input signals corresponding to a user's entire speech can be processed and converted into recognizable information, then the speech recognition system can process the entire audio input signals and convert it to recognizable information at step 206 , without initiating the video signal speech recognition functions.
  • the speech recognition system can implement one or more task(s) based on the recognizable information at step 207 .
  • the speaker can talk into a cell phone configured with the speech recognition system of the present invention.
  • the speaker can request to dial a particular number or connect with the Internet. Therefore, the speech recognition system can convert the speech into either recognizable information such as numeric characters like dialing a particular number, or convert the speech into recognizable information such as a set of code(s) to perform a particular function like connecting with the Internet.
  • the audio signal speech recognition processing functions can become the primary processing functions of the speech recognition system until a section of the audio input signals cannot be processed and converted into recognizable information.
  • the present invention can access the correlating portion of the video input signals at step 208 to assist in recognizing the speech.
  • the speech recognition system can access a portion of the video image of the lip movements of the speaker, wherein the video image can correspond to the unrecognizable audio signal portion.
  • audio input signals based on a user's speech can be received by the speech recognition system, and when the audio speech recognition portion of the system detects a possible conversion error that is equal to or is above a predetermined threshold level, then the video speech recognition portion of the system can be initiated.
  • the system can access the video images of the lip movements correlating to the audio speech in question and can determine the movements of the lips at step 209 .
  • the system can thereafter process the video images and assist in the conversion of audio and video input signals to recognizable and comprehensible information at step 210 .
  • the video input signals can be processed to assist in speech recognition, the video input signal can also be processed not just as an aid to the audio input signals but as a stand-alone speech recognition feature of the system.
  • the speech recognition system can implement a task based on the recognizable information at step 211 .
  • the combination of both the audio speech and the video image of the lip movement can resolve unrecognizable audio speech.
  • the system can be configured to identify likely sounds corresponding to certain lip movements, and can also be configured to recognize speech based on the context in which the word or sound was spoken.
  • the present invention can resolve unrecognizable speech by referring to the adjacent recognizable words or phrases within the speech in order to aid in the recognition of the unrecognizable portion of the speech.
  • FIGS. 3 and 4 illustrates one example of a method of sending the audio and video input signals to a destination site where the audio and video input signals can be processed and converted into recognizable and comprehensible information.
  • FIGS. 3 and 4 illustrate one example of a method of speech recognition at a destination site using audio input signals together with correlating video images of lip movements.
  • the method of the present example can be implemented in hardware, or software, or a combination of both hardware and software.
  • a device configured to include a speech recognition system can be activated at step 300 of FIG. 3 .
  • a detecting sensor along with an optical pick-up such as a lens, can detect for video images resembling the user's lips at step 301 . If the detecting sensor and the lens do not detect images of the user's lips, then the speech recognition system can alert the user to readjust the lens or the device, or reposition the user's lips so that an image can be detected, at step 302 of FIG. 3 . If however, the user's lips can be detected or captured by the lens and the sensor, then the speech recognition system can alert the speaker in step 303 that the speech recognition system is in ready mode, and therefore the user can commence speaking.
  • the speech recognition system of the present invention can commence receiving both audio and video input signals from the user's speech and the user's lip movements at step 304 .
  • the received audio input signals and the received video input signals can be stored within a storage unit and/or a plurality of separate storage units.
  • the speech recognition system can detect, based on sensors and preprogrammed conditions, an end of speech status at step 305 .
  • speech recognition system can activate an end of speech condition. Thereafter, the speech recognition system can prompt the user if the user desires to send the stored speech at step 306 .
  • the stored speech can remain stored in the storage unit(s) and can be recalled at later time. However, if the user responds in the positive, then the speech recognition system can transmit the stored speech to a destination site at step 307 .
  • the destination site can activate the audio and video speech recognition system available at the destination site at step 400 . Thereafter, the audio and video speech recognition system can process and convert the audio and video signals to recognizable and comprehensible information as discussed above with respect to FIG. 2 .
  • the speech recognition system at the destination site can preliminary determine if the audio input signal can be processed and converted to recognizable information at step 401 . If the entire audio portion of the speech or the entire audio input signals can be processed and converted as recognizable information, then the system can do so without activating the video speech recognition portion of the system at step 406 . Thus, the entire audio portion of the speech can be processed and converted into recognizable information, and the speech recognition system can implement one or more task(s) based on the recognizable information at step 407 .
  • the present invention can access the correlating portion of the video input signals at step 402 to assist in recognizing the speech.
  • the speech recognition system can access a portion of the video image of the lip movements of the speaker, wherein the video image can correspond to the unrecognizable audio signal portion.
  • the system can access the video images of the lip movements correlating to the audio speech in question and can process and determine the movements of the lips at step 403 .
  • the system can thereafter process the video images and assist in the conversion of audio and video input signals to recognizable and comprehensible information at step 404 .
  • the speech recognition system can implement one or more task(s) based on the recognizable information at step 405 .
  • the speech recognition system of the present invention can simultaneously process and convert the audio input signals and the video input signals in parallel.
  • the system can initiate both the audio speech recognition in tandem with the video speech recognition. Therefore, the speech recognition system can process the audio input signals and the correlating video input signals in parallel, and can convert the audio speech and the correlating video images of the lip movements into recognizable and comprehensible information.
  • FIG. 5 illustrates one example of a hardware configuration that can perform speech recognition based on audio input signals and correlating video input signals, in accordance with the present invention.
  • the hardware configuration of FIG. 5 can be in an integrated, modular and single chip solution, and therefore can be embodied on a semiconductor substrate, such as silicon.
  • the hardware configuration of FIG. 5 can be a plurality of discrete components on a circuit board.
  • the configuration can also be implemented as a general purpose device configured to implement the invention with software.
  • FIG. 5 illustrates a device 500 configured to perform speech recognition based on audio signals and correlating video images of lip movements.
  • Device 500 can contain an audio receiving unit 505 and a video receiving unit 510 .
  • the audio receiving unit 505 can receive audio input signals from one or more audio source(s) such as voice, speech, music, etc.
  • the video receiving unit 510 can receive video input signals from one or more video source(s).
  • the video receiving unit 510 can receive video images of a speaker's lip movements.
  • the device 500 can include a video image sensor 515 , wherein the sensor 515 can detect when a particular image such as a speaker's lips is not being received by the video receiving unit 510 . In other words, if the speaker's lips are not positioned in a way for the video receiving unit to receive video images of lips, then the sensor can detect missing video images and can alert the speaker.
  • the device 500 can include a processing unit 520 and a converting unit 525 .
  • the processing unit 520 can process the audio input signals as well as the video input signals.
  • the converting unit 525 can convert the processed audio input signals and the video input signals into recognizable and comprehensible information. For instance, the converting unit 525 can convert the processed audio input signals and the video images of lip movements into executable commands or into text, etc. If the converted signals are commands to perform one or a set of function(s), then the implementation unit 530 can execute the command(s).

Abstract

A speech recognition device can include an audio signal receiver configured to receive audio signals from a speech source, a video signal receiver configured to receive video signals from the speech source, and a processing unit configured to process the audio signals and the video signals. In addition, the speech recognition device can include a conversion unit configured to convert the audio signals and the video signals to recognizable speech, and an implementation unit configured to implement a task based on the recognizable speech.

Description

    CROSS REFERENCE TO RELATED APPLICATIONS
  • This application claims priority of U.S. Provisional Patent Application Ser. No. 60/409,956, filed Sep. 12, 2002, and 60/445,816, filed Feb. 10, 2003, entitled Correlating Video Images of Lip Movements with Audio Signals to Improve Speech Recognition. The contents of the provisional applications are hereby incorporated by reference.
  • BACKGROUND OF THE INVENTION
  • 1. Field of the Invention
  • The present invention relates to a method of and an apparatus for using video signals along with audio signals of speech to provide speech recognition, within an environment where speech recognition is necessary. In particular, the present invention relates to a method of and a system for using video images of lip movements with audio input signals to improve speech recognition. The present invention can be implemented in a hand held device, and the invention may include discrete devices or may be implemented on a semiconductor substrate such as a silicon chip.
  • 2. Description of the Related Art
  • Human speech is made up of numerous different sounds and syllables. Often in many languages, different sounds and/or syllables are combined to form words and/or sentences. The combination of the sounds, syllables, words, and sentences forms the basis for oral communication.
  • Generally, human speech is recognizable if the speech is clear and comprehensible to another human's ears. On the other hand, human speech can be recognizable by a machine if the audio waves of the speech is received, and the audio waves are recognizable by an algorithm operating within the machine. Although audio speech recognition by machines has advanced in sophistication, the accuracy of audio speech recognition has room for improvements.
  • SUMMARY OF THE INVENTION
  • One example of the present invention can be a method of speech recognition. The method can include the steps of receiving audio signals from, a speech source, receiving video signals from the speech source, and processing the audio signals and the video signals. The method can also include the steps of converting the audio signals and the video signals to recognizable information, and implementing a task based on the recognizable information.
  • In another example, the present invention can relate to a speech recognition device. The device can have an audio signal receiver configured to receive audio signals from a speech source, a video signal receiver configured to receive video signals from the speech source, and a processing unit configured to process the audio signals and the video signals. Moreover, the device can have a conversion unit configured to convert the audio signals and the video signals to recognizable information, and an implementation unit configured to implement a task based on the recognizable information.
  • Additionally, another example of the present invention can provide a system for speech recognition. The system can include a first receiving means for receiving audio signals from a speech source, a second receiving means for receiving video signals from the speech source, and a processing means for processing the audio signals and the video signals. Furthermore, the system can have a converting means for converting the audio signals and the video signals to recognizable information, and an implementing means for implementing a task based on the recognizable information.
  • Furthermore, another example of the present invention can be directed to a method of speech recognition. The method can include the steps of receiving audio signals from a speech source, receiving video signals from the speech source, processing the audio signals, and converting the audio signals into recognizable information. Moreover, the method can have the step of processing the video signals when a segment of the audio signals can not be converted into the recognizable information. The video signals can coincide with the segment of the audio signals that cannot be converted into the recognizable information. The method also can have the steps of converting the processed video signals into the recognizable information, and implementing a task based on the recognizable information.
  • In another example, the present invention can be a speech recognition device. The device can have an audio signal receiver configured to receive audio signals from a speech source, a video signal receiver configured to receive video signals from the speech source, a first processing unit configured to process the audio signals, and a first conversion unit configured to convert the audio signals to recognizable information. The device can also have a second processing unit configured to process the video signals when the audio signals cannot be converted into the recognizable information, wherein the video signals coincide with the segment of the audio signals that cannot be converted into the recognizable information, a second conversion unit configured to convert the video signals processed into the recognizable information, and an implementation unit configured to implement a task based on the recognizable information.
  • In yet another example, the present invention can be drawn to a system for speech recognition. The system can include a first receiving means for receiving audio signals from a speech source, a second receiving means for receiving video signals from the speech source, a first processing means for processing the audio signals, and a first converting means for converting the audio signals into recognizable information. The system can also have a second processing means for processing the video signals when a segment of the audio signals can not be converted into the recognizable information, wherein the video signals coincide with the segment of the audio signals that cannot be converted into the recognizable information, a second converting means for converting the video signals processed into the recognizable information, and an implementing means for implementing a task based on the recognizable information.
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • For proper understanding of the invention, reference should be made to the accompanying drawings, wherein:
  • FIG. 1 illustrates one example of a speech recognition device using audio signals and correlating video signals to improve speech recognition, in accordance with the present invention;
  • FIG. 2 illustrates a flow chart illustrating one example of a method of speech recognition using audio signals and correlating video signals, in accordance with the present invention;
  • FIG. 3 illustrates a flow chart illustrating another example of a method of speech recognition using audio signals and correlating video signals, in accordance with the present invention;
  • FIG. 4 illustrates a flow chart of a method of speech recognition using audio signals and correlating video signals, in accordance with the present invention; and
  • FIG. 5 illustrates one example of a hardware configuration for speech recognition using audio signals and correlating video signals, in accordance with the present invention.
  • DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT(S):
  • FIG. 1 illustrates one example of a speech recognition device for improving speech recognition with audio input signals and correlating video input signals according to the present invention. FIG. 1 shows a mobile phone 100 having a display screen 101 and a plurality of actuators 102. In addition, the mobile phone 100 can include a lens 103 for receiving images in video and/or still picture format. In the example shown in FIG. 1, the lens 103 can capture or receive video images of the movements of a user's lips 104 a, 104 b made while the user is speaking.
  • For instance, a user may desire to contact a business associate via an e-mail message. According, the user can access the mobile phone 100 and can activate the speech recognition system of the present invention. The mobile phone 100 is placed adjacent to the user's face where the lens 103 is positioned in proximity to the user's mouth so that the image of the user's lips can be captured by the speech recognition system. Once the lens 103 is positioned correctly, the user can be alerted to commence speaking into the mobile phone 100. The user, for example, can speak and request to send an e-mail message to Jane Doe in which her e-mail address is pre-programmed into the mobile phone 100. In addition, the name James Doe is similarly pre-programmed in the mobile phone 100. The speech recognition system processes the audio signal input and can convert all part of the audio signal input into recognizable information with the exception of the name Jane Doe. The audio speech recognition feature of the invention cannot ascertain if the audio signal is referring to Jane Doe or James Doe.
  • Accordingly, the speech recognition system of the present invention can access a section of the video input signal corresponding to the audio signal pertaining to Jane Doe. Based on the detected lip movements of the video signals, the present invention can reduce the uncertainty of the audio speech recognition, and therefore can perform speech recognition with the aid of the video signals, and determine that the request is an e-mail message to Jane Doe rather than James Doe. Thereafter, the present invention can implement one or more actions to carry out the spoken request by the user. The one or more actions can be in the forms of commands such as initiating an e-mail application software, creating an outgoing e-mail window, and inserting text, and sending the e-mail message.
  • Although the example provided in FIG. 1 illustrates a mobile phone 100 having a lens 103, wherein the mobile phone 100 can be configured with the speech recognition system of the present invention, it is noted that the speech recognition system using audio signals and correlating video signals of the invention can be configured on a variety of electronic device, either mobile or stationary. For instance, the improved speech recognition system of the invention can be configured on at least but not limited to a laptop computer, a PDA, an audio/video recording device, a home computer, a game console, a remote controller, or other comparable device.
  • FIG. 2 illustrates one example of a method of speech recognition using audio input signal and correlating video input signals, in accordance with the present invention. Specifically, FIG. 2 illustrates one example of a method of speech recognition using audio input signals together with correlating video images of lip movements. The method of the present example can be implemented in hardware, or software, or a combination of both hardware and software.
  • FIG. 2 illustrates one example of a method of speech recognition according to the present invention. A device configured to include a speech recognition system can be activated at step 200 of FIG. 2. In other words, the present invention provides a user with the option to activate the speech recognition feature when necessary. After the speech recognition system is activated, a detecting sensor along with an optical pick-up such as a lens, can detect for video images resembling the user's lips at step 201. If the detecting sensor and the lens do not detect images of the user's lips, then the speech recognition system can alert the user to readjust the lens or the device, or reposition the user's lips so that an image can be detected, at step 202 of FIG. 2. If however, the user's lips can be detected or captured by the lens and the sensor, then the speech recognition system can alert the speaker in step 203 that the speech recognition system is in ready mode, and therefore the user can commence speaking.
  • Once the user starts to speak, the speech recognition system of the present invention can commence receiving both audio and video input signals from the user's speech and the user's lip movements at step 204. In this example, the speech recognition system can process the audio input signals corresponding to the speech first. In other words, as the user speaks, both the audio speech and the correlating images of the user's lip movements can be received by the speech recognition system of a device. Although both audio and video signals are being received, the speech recognition system can preliminarily initiate only the audio speech recognition portion of the system, and can preliminarily process only the audio portion of the speech.
  • Therefore, if the speech from the user does not contain a possibly unrecognizable sound or word, then the present invention at step 206 can recognize the speech as comprehensible and recognizable information using only the audio speech recognition portion of the system without the need to activate the assistance of the video signals.
  • The speech recognition system can process the audio input signals and determine if the audio input signals are recognizable as speech at step 205. If it is determined that the audio input signals corresponding to a user's entire speech can be processed and converted into recognizable information, then the speech recognition system can process the entire audio input signals and convert it to recognizable information at step 206, without initiating the video signal speech recognition functions.
  • Thereafter, the speech recognition system can implement one or more task(s) based on the recognizable information at step 207. For instance, the speaker can talk into a cell phone configured with the speech recognition system of the present invention. The speaker can request to dial a particular number or connect with the Internet. Therefore, the speech recognition system can convert the speech into either recognizable information such as numeric characters like dialing a particular number, or convert the speech into recognizable information such as a set of code(s) to perform a particular function like connecting with the Internet. Accordingly, the audio signal speech recognition processing functions can become the primary processing functions of the speech recognition system until a section of the audio input signals cannot be processed and converted into recognizable information.
  • If however a section of the audio input signals cannot be process and converted into recognizable information, the present invention can access the correlating portion of the video input signals at step 208 to assist in recognizing the speech. In other words, whenever the audio speech recognition portion of the system identifies a possibly unrecognizable sound or word, then the speech recognition system can access a portion of the video image of the lip movements of the speaker, wherein the video image can correspond to the unrecognizable audio signal portion. For instance, audio input signals based on a user's speech can be received by the speech recognition system, and when the audio speech recognition portion of the system detects a possible conversion error that is equal to or is above a predetermined threshold level, then the video speech recognition portion of the system can be initiated.
  • Once the video speech recognition portion of the system is initiated, the system can access the video images of the lip movements correlating to the audio speech in question and can determine the movements of the lips at step 209. The system can thereafter process the video images and assist in the conversion of audio and video input signals to recognizable and comprehensible information at step 210. It is noted that although the video input signals can be processed to assist in speech recognition, the video input signal can also be processed not just as an aid to the audio input signals but as a stand-alone speech recognition feature of the system.
  • Following the processing and converting of the video input signals correlating to the audio signal in question, the speech recognition system can implement a task based on the recognizable information at step 211.
  • Thus, the combination of both the audio speech and the video image of the lip movement can resolve unrecognizable audio speech. In addition, the system can be configured to identify likely sounds corresponding to certain lip movements, and can also be configured to recognize speech based on the context in which the word or sound was spoken. In other words, the present invention can resolve unrecognizable speech by referring to the adjacent recognizable words or phrases within the speech in order to aid in the recognition of the unrecognizable portion of the speech.
  • In an alternative example, the present invention can recognize speech using both audio and video signals at a destination site rather than at the originator. FIGS. 3 and 4 illustrates one example of a method of sending the audio and video input signals to a destination site where the audio and video input signals can be processed and converted into recognizable and comprehensible information. Specifically, FIGS. 3 and 4 illustrate one example of a method of speech recognition at a destination site using audio input signals together with correlating video images of lip movements. The method of the present example can be implemented in hardware, or software, or a combination of both hardware and software.
  • A device configured to include a speech recognition system can be activated at step 300 of FIG. 3. After the speech recognition system is activated, a detecting sensor along with an optical pick-up such as a lens, can detect for video images resembling the user's lips at step 301. If the detecting sensor and the lens do not detect images of the user's lips, then the speech recognition system can alert the user to readjust the lens or the device, or reposition the user's lips so that an image can be detected, at step 302 of FIG. 3. If however, the user's lips can be detected or captured by the lens and the sensor, then the speech recognition system can alert the speaker in step 303 that the speech recognition system is in ready mode, and therefore the user can commence speaking.
  • Once the user starts to speak, the speech recognition system of the present invention can commence receiving both audio and video input signals from the user's speech and the user's lip movements at step 304. The received audio input signals and the received video input signals can be stored within a storage unit and/or a plurality of separate storage units.
  • Following the completion of the user's speech, the speech recognition system can detect, based on sensors and preprogrammed conditions, an end of speech status at step 305. In other word, once the sensors detects that the user has completed his speech and that certain preprogrammed conditions have been met, then speech recognition system can activate an end of speech condition. Thereafter, the speech recognition system can prompt the user if the user desires to send the stored speech at step 306.
  • If the user responds in the negative, then the stored speech can remain stored in the storage unit(s) and can be recalled at later time. However, if the user responds in the positive, then the speech recognition system can transmit the stored speech to a destination site at step 307.
  • After the stored speech is received at the destination site, then the destination site can activate the audio and video speech recognition system available at the destination site at step 400. Thereafter, the audio and video speech recognition system can process and convert the audio and video signals to recognizable and comprehensible information as discussed above with respect to FIG. 2. In other words, the speech recognition system at the destination site can preliminary determine if the audio input signal can be processed and converted to recognizable information at step 401. If the entire audio portion of the speech or the entire audio input signals can be processed and converted as recognizable information, then the system can do so without activating the video speech recognition portion of the system at step 406. Thus, the entire audio portion of the speech can be processed and converted into recognizable information, and the speech recognition system can implement one or more task(s) based on the recognizable information at step 407.
  • If, however, a section of the audio input signals cannot be process and converted into recognizable information, the present invention can access the correlating portion of the video input signals at step 402 to assist in recognizing the speech. In other words, whenever the audio speech recognition portion of the system detects a possibly unrecognizable sound or word, then the speech recognition system can access a portion of the video image of the lip movements of the speaker, wherein the video image can correspond to the unrecognizable audio signal portion.
  • Once the video speech recognition portion of the system is triggered, the system can access the video images of the lip movements correlating to the audio speech in question and can process and determine the movements of the lips at step 403. The system can thereafter process the video images and assist in the conversion of audio and video input signals to recognizable and comprehensible information at step 404. After the conversion of the audio and/or video input signals, the speech recognition system can implement one or more task(s) based on the recognizable information at step 405.
  • It is noted that the speech recognition system of the present invention can simultaneously process and convert the audio input signals and the video input signals in parallel. In other words, rather than the system initiating the audio speech recognition portion of the system to first process and convert the audio portion of the speech, the system can initiate both the audio speech recognition in tandem with the video speech recognition. Therefore, the speech recognition system can process the audio input signals and the correlating video input signals in parallel, and can convert the audio speech and the correlating video images of the lip movements into recognizable and comprehensible information.
  • FIG. 5 illustrates one example of a hardware configuration that can perform speech recognition based on audio input signals and correlating video input signals, in accordance with the present invention. In addition, the hardware configuration of FIG. 5 can be in an integrated, modular and single chip solution, and therefore can be embodied on a semiconductor substrate, such as silicon. Alternatively, the hardware configuration of FIG. 5 can be a plurality of discrete components on a circuit board. The configuration can also be implemented as a general purpose device configured to implement the invention with software.
  • FIG. 5 illustrates a device 500 configured to perform speech recognition based on audio signals and correlating video images of lip movements. Device 500 can contain an audio receiving unit 505 and a video receiving unit 510. The audio receiving unit 505 can receive audio input signals from one or more audio source(s) such as voice, speech, music, etc. The video receiving unit 510 can receive video input signals from one or more video source(s). For example, the video receiving unit 510 can receive video images of a speaker's lip movements. In addition, the device 500 can include a video image sensor 515, wherein the sensor 515 can detect when a particular image such as a speaker's lips is not being received by the video receiving unit 510. In other words, if the speaker's lips are not positioned in a way for the video receiving unit to receive video images of lips, then the sensor can detect missing video images and can alert the speaker.
  • Furthermore, the device 500 can include a processing unit 520 and a converting unit 525. The processing unit 520 can process the audio input signals as well as the video input signals. The converting unit 525 can convert the processed audio input signals and the video input signals into recognizable and comprehensible information. For instance, the converting unit 525 can convert the processed audio input signals and the video images of lip movements into executable commands or into text, etc. If the converted signals are commands to perform one or a set of function(s), then the implementation unit 530 can execute the command(s).
  • One having ordinary skill in the art will readily understand that the invention as discussed above may be practiced with steps in a different order, and/or with hardware elements in configurations which are different than those which are disclosed. Therefore, although the invention has been described based upon these preferred embodiments, it would be apparent to those of skill in the art that certain modifications, variations, and alternative constructions would be apparent, while remaining within the spirit and scope of the invention. In order to determine the metes and bounds of the invention, therefore, reference should be made to the appended claims.

Claims (2)

1. A method of speech recognition, said method comprising the steps of: receiving audio signals from a speech source;
receiving video signals from the speech source;
processing the audio signals and the video signals;
converting the audio signals and the video signals into recognizable information;
implementing a task based on the recognizable information.
2-15. (canceled)
US12/553,663 2002-09-12 2009-09-03 Correlating video images of lip movements with audio signals to improve speech recognition Abandoned US20100063820A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US12/553,663 US20100063820A1 (en) 2002-09-12 2009-09-03 Correlating video images of lip movements with audio signals to improve speech recognition

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US40995602P 2002-09-12 2002-09-12
US44581603P 2003-02-10 2003-02-10
US10/660,780 US7587318B2 (en) 2002-09-12 2003-09-12 Correlating video images of lip movements with audio signals to improve speech recognition
US12/553,663 US20100063820A1 (en) 2002-09-12 2009-09-03 Correlating video images of lip movements with audio signals to improve speech recognition

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
US10/660,780 Continuation US7587318B2 (en) 2002-09-12 2003-09-12 Correlating video images of lip movements with audio signals to improve speech recognition

Publications (1)

Publication Number Publication Date
US20100063820A1 true US20100063820A1 (en) 2010-03-11

Family

ID=32512290

Family Applications (2)

Application Number Title Priority Date Filing Date
US10/660,780 Expired - Fee Related US7587318B2 (en) 2002-09-12 2003-09-12 Correlating video images of lip movements with audio signals to improve speech recognition
US12/553,663 Abandoned US20100063820A1 (en) 2002-09-12 2009-09-03 Correlating video images of lip movements with audio signals to improve speech recognition

Family Applications Before (1)

Application Number Title Priority Date Filing Date
US10/660,780 Expired - Fee Related US7587318B2 (en) 2002-09-12 2003-09-12 Correlating video images of lip movements with audio signals to improve speech recognition

Country Status (1)

Country Link
US (2) US7587318B2 (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110257971A1 (en) * 2010-04-14 2011-10-20 T-Mobile Usa, Inc. Camera-Assisted Noise Cancellation and Speech Recognition
WO2016082267A1 (en) * 2014-11-28 2016-06-02 深圳创维-Rgb电子有限公司 Voice recognition method and system
WO2016150001A1 (en) * 2015-03-24 2016-09-29 中兴通讯股份有限公司 Speech recognition method, device and computer storage medium
US9870500B2 (en) 2014-06-11 2018-01-16 At&T Intellectual Property I, L.P. Sensor enhanced speech recognition
TWI662542B (en) * 2018-04-26 2019-06-11 鴻海精密工業股份有限公司 Voice recognition device and method
CN110415689A (en) * 2018-04-26 2019-11-05 富泰华工业(深圳)有限公司 Speech recognition equipment and method
US11341962B2 (en) 2010-05-13 2022-05-24 Poltorak Technologies Llc Electronic personal interactive device
US11818458B2 (en) 2005-10-17 2023-11-14 Cutting Edge Vision, LLC Camera touchpad

Families Citing this family (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7587318B2 (en) * 2002-09-12 2009-09-08 Broadcom Corporation Correlating video images of lip movements with audio signals to improve speech recognition
US20070124147A1 (en) * 2005-11-30 2007-05-31 International Business Machines Corporation Methods and apparatus for use in speech recognition systems for identifying unknown words and for adding previously unknown words to vocabularies and grammars of speech recognition systems
US7787697B2 (en) * 2006-06-09 2010-08-31 Sony Ericsson Mobile Communications Ab Identification of an object in media and of related media objects
US20100079573A1 (en) * 2008-09-26 2010-04-01 Maycel Isaac System and method for video telephony by converting facial motion to text
KR101092820B1 (en) * 2009-09-22 2011-12-12 현대자동차주식회사 Lipreading and Voice recognition combination multimodal interface system
JP2011186351A (en) * 2010-03-11 2011-09-22 Sony Corp Information processor, information processing method, and program
US8751228B2 (en) 2010-11-04 2014-06-10 Microsoft Corporation Minimum converted trajectory error (MCTE) audio-to-video engine
US8863042B2 (en) * 2012-01-24 2014-10-14 Charles J. Kulas Handheld device with touch controls that reconfigure in response to the way a user operates the device
US9263044B1 (en) * 2012-06-27 2016-02-16 Amazon Technologies, Inc. Noise reduction based on mouth area movement recognition
WO2014025012A1 (en) * 2012-08-10 2014-02-13 株式会社ホンダアクセス Speech recognition method and speech recognition device
KR101442211B1 (en) * 2013-02-07 2014-10-16 서강대학교산학협력단 Speech recognition system and method using 3D geometric information
US9264824B2 (en) * 2013-07-31 2016-02-16 Starkey Laboratories, Inc. Integration of hearing aids with smart glasses to improve intelligibility in noise
US20150084859A1 (en) * 2013-09-23 2015-03-26 Yair ITZHAIK System and Method for Recognition and Response to Gesture Based Input
US20150149169A1 (en) * 2013-11-27 2015-05-28 At&T Intellectual Property I, L.P. Method and apparatus for providing mobile multimodal speech hearing aid
CN105991851A (en) 2015-02-17 2016-10-05 杜比实验室特许公司 Endpoint device for processing disturbance in telephone conference system
US9940932B2 (en) 2016-03-02 2018-04-10 Wipro Limited System and method for speech-to-text conversion
JP2018091954A (en) * 2016-12-01 2018-06-14 オリンパス株式会社 Voice recognition device and voice recognition method
US10332515B2 (en) * 2017-03-14 2019-06-25 Google Llc Query endpointing based on lip detection
US10719692B2 (en) 2017-09-09 2020-07-21 Apple Inc. Vein matching for difficult biometric authentication cases
US10522147B2 (en) * 2017-12-21 2019-12-31 Motorola Solutions, Inc. Device and method for generating text representative of lip movement
KR102629385B1 (en) * 2018-01-25 2024-01-25 삼성전자주식회사 Application processor including low power voice trigger system with direct path for barge-in, electronic device including the same and method of operating the same
EP3752957A4 (en) * 2018-02-15 2021-11-17 DMAI, Inc. System and method for speech understanding via integrated audio and visual based speech recognition
WO2019161196A2 (en) * 2018-02-15 2019-08-22 DMAI, Inc. System and method for disambiguating a source of sound based on detected lip movement
US11468885B2 (en) 2018-02-15 2022-10-11 DMAI, Inc. System and method for conversational agent via adaptive caching of dialogue tree
US11308312B2 (en) 2018-02-15 2022-04-19 DMAI, Inc. System and method for reconstructing unoccupied 3D space
US11069357B2 (en) 2019-07-31 2021-07-20 Ebay Inc. Lip-reading session triggering events
US11244696B2 (en) 2019-11-06 2022-02-08 Microsoft Technology Licensing, Llc Audio-visual speech enhancement
CN111161724B (en) * 2019-12-16 2022-12-13 爱驰汽车有限公司 Method, system, equipment and medium for Chinese audio-visual combined speech recognition
CN113002461A (en) * 2021-03-26 2021-06-22 芜湖汽车前瞻技术研究院有限公司 Virtual image position adjusting method, device and storage medium of AR-HUD system

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4769845A (en) * 1986-04-10 1988-09-06 Kabushiki Kaisha Carrylab Method of recognizing speech using a lip image
US5313522A (en) * 1991-08-23 1994-05-17 Slager Robert P Apparatus for generating from an audio signal a moving visual lip image from which a speech content of the signal can be comprehended by a lipreader
US20020035475A1 (en) * 2000-09-12 2002-03-21 Pioneer Corporation Voice recognition apparatus
US20020116197A1 (en) * 2000-10-02 2002-08-22 Gamze Erten Audio visual speech processing
US6471420B1 (en) * 1994-05-13 2002-10-29 Matsushita Electric Industrial Co., Ltd. Voice selection apparatus voice response apparatus, and game apparatus using word tables from which selected words are output as voice selections
US20030018475A1 (en) * 1999-08-06 2003-01-23 International Business Machines Corporation Method and apparatus for audio-visual speech detection and recognition
US6754373B1 (en) * 2000-07-14 2004-06-22 International Business Machines Corporation System and method for microphone activation using visual speech cues
US20050228673A1 (en) * 2004-03-30 2005-10-13 Nefian Ara V Techniques for separating and evaluating audio and video source data
US7587318B2 (en) * 2002-09-12 2009-09-08 Broadcom Corporation Correlating video images of lip movements with audio signals to improve speech recognition
US8078469B2 (en) * 1999-04-12 2011-12-13 White George M Distributed voice user interface

Family Cites Families (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4786987A (en) * 1985-10-19 1988-11-22 Fuji Photo Film Co., Ltd. Electronic camera with autofocus warning device
US4891660A (en) * 1988-11-29 1990-01-02 Pvi, Inc. Automatic photographic system and frame dispenser
US5271011A (en) * 1992-03-16 1993-12-14 Scientific-Atlanta, Inc. Digital audio data muting system and method
US5502774A (en) * 1992-06-09 1996-03-26 International Business Machines Corporation Automatic recognition of a consistent message using multiple complimentary sources of information
IT1257073B (en) * 1992-08-11 1996-01-05 Ist Trentino Di Cultura RECOGNITION SYSTEM, ESPECIALLY FOR THE RECOGNITION OF PEOPLE.
US5761329A (en) * 1995-12-15 1998-06-02 Chen; Tsuhan Method and apparatus employing audio and video data from an individual for authentication purposes
US6119083A (en) * 1996-02-29 2000-09-12 British Telecommunications Public Limited Company Training process for the classification of a perceptual signal
US6219639B1 (en) * 1998-04-28 2001-04-17 International Business Machines Corporation Method and apparatus for recognizing identity of individuals employing synchronized biometrics
JP3893763B2 (en) * 1998-08-17 2007-03-14 富士ゼロックス株式会社 Voice detection device
US6366296B1 (en) * 1998-09-11 2002-04-02 Xerox Corporation Media browser using multimodal analysis
US6243683B1 (en) * 1998-12-29 2001-06-05 Intel Corporation Video control of speech recognition
US6583821B1 (en) * 1999-07-16 2003-06-24 Thomson Licensing S.A. Synchronizing apparatus for a compressed audio/video signal receiver
US6219640B1 (en) * 1999-08-06 2001-04-17 International Business Machines Corporation Methods and apparatus for audio-visual speaker recognition and utterance verification
JP3485508B2 (en) * 1999-10-26 2004-01-13 株式会社国際電気通信基礎技術研究所 Facial image transmitting method and system, and facial image transmitting device and facial image reproducing device used in the system
US6633844B1 (en) * 1999-12-02 2003-10-14 International Business Machines Corporation Late integration in audio-visual continuous speech recognition
US6526395B1 (en) * 1999-12-31 2003-02-25 Intel Corporation Application of personality models and interaction with synthetic characters in a computing system
EP1215660B1 (en) * 2000-12-14 2004-03-10 TELEFONAKTIEBOLAGET L M ERICSSON (publ) Mobile terminal controllable by spoken utterances
US20020077830A1 (en) * 2000-12-19 2002-06-20 Nokia Corporation Method for activating context sensitive speech recognition in a terminal
US7271839B2 (en) * 2001-03-15 2007-09-18 Lg Electronics Inc. Display device of focal angle and focal distance in iris recognition system
US6931351B2 (en) * 2001-04-20 2005-08-16 International Business Machines Corporation Decision making in classification problems
US7076429B2 (en) * 2001-04-27 2006-07-11 International Business Machines Corporation Method and apparatus for presenting images representative of an utterance with corresponding decoded speech
US7069215B1 (en) * 2001-07-12 2006-06-27 At&T Corp. Systems and methods for extracting meaning from multimodal inputs using finite-state devices
US6868383B1 (en) * 2001-07-12 2005-03-15 At&T Corp. Systems and methods for extracting meaning from multimodal inputs using finite-state devices
US7343082B2 (en) * 2001-09-12 2008-03-11 Ryshco Media Inc. Universal guide track
US7019791B2 (en) * 2001-11-09 2006-03-28 Hitachi, Ltd. Video processing device
US6950536B2 (en) * 2002-01-25 2005-09-27 Houvener Robert C High volume mobile identity verification system and method using tiered biometric analysis
US7219062B2 (en) * 2002-01-30 2007-05-15 Koninklijke Philips Electronics N.V. Speech activity detection using acoustic and facial characteristics in an automatic speech recognition system
US7165029B2 (en) * 2002-05-09 2007-01-16 Intel Corporation Coupled hidden Markov model for audiovisual speech recognition
US20030212552A1 (en) * 2002-05-09 2003-11-13 Liang Lu Hong Face recognition procedure useful for audiovisual speech recognition
US7251603B2 (en) * 2003-06-23 2007-07-31 International Business Machines Corporation Audio-only backoff in audio-visual speech recognition system
US7269560B2 (en) * 2003-06-27 2007-09-11 Microsoft Corporation Speech detection and enhancement using audio/video fusion
US7283850B2 (en) * 2004-10-12 2007-10-16 Microsoft Corporation Method and apparatus for multi-sensory speech enhancement on a mobile device
US7518631B2 (en) * 2005-06-28 2009-04-14 Microsoft Corporation Audio-visual control system

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4769845A (en) * 1986-04-10 1988-09-06 Kabushiki Kaisha Carrylab Method of recognizing speech using a lip image
US5313522A (en) * 1991-08-23 1994-05-17 Slager Robert P Apparatus for generating from an audio signal a moving visual lip image from which a speech content of the signal can be comprehended by a lipreader
US6471420B1 (en) * 1994-05-13 2002-10-29 Matsushita Electric Industrial Co., Ltd. Voice selection apparatus voice response apparatus, and game apparatus using word tables from which selected words are output as voice selections
US8078469B2 (en) * 1999-04-12 2011-12-13 White George M Distributed voice user interface
US20030018475A1 (en) * 1999-08-06 2003-01-23 International Business Machines Corporation Method and apparatus for audio-visual speech detection and recognition
US6754373B1 (en) * 2000-07-14 2004-06-22 International Business Machines Corporation System and method for microphone activation using visual speech cues
US20020035475A1 (en) * 2000-09-12 2002-03-21 Pioneer Corporation Voice recognition apparatus
US20020116197A1 (en) * 2000-10-02 2002-08-22 Gamze Erten Audio visual speech processing
US7587318B2 (en) * 2002-09-12 2009-09-08 Broadcom Corporation Correlating video images of lip movements with audio signals to improve speech recognition
US20050228673A1 (en) * 2004-03-30 2005-10-13 Nefian Ara V Techniques for separating and evaluating audio and video source data

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11818458B2 (en) 2005-10-17 2023-11-14 Cutting Edge Vision, LLC Camera touchpad
US8635066B2 (en) * 2010-04-14 2014-01-21 T-Mobile Usa, Inc. Camera-assisted noise cancellation and speech recognition
US20110257971A1 (en) * 2010-04-14 2011-10-20 T-Mobile Usa, Inc. Camera-Assisted Noise Cancellation and Speech Recognition
US11341962B2 (en) 2010-05-13 2022-05-24 Poltorak Technologies Llc Electronic personal interactive device
US11367435B2 (en) 2010-05-13 2022-06-21 Poltorak Technologies Llc Electronic personal interactive device
US9870500B2 (en) 2014-06-11 2018-01-16 At&T Intellectual Property I, L.P. Sensor enhanced speech recognition
US10083350B2 (en) 2014-06-11 2018-09-25 At&T Intellectual Property I, L.P. Sensor enhanced speech recognition
WO2016082267A1 (en) * 2014-11-28 2016-06-02 深圳创维-Rgb电子有限公司 Voice recognition method and system
US20170098447A1 (en) * 2014-11-28 2017-04-06 Shenzhen Skyworth-Rgb Electronic Co., Ltd. Voice recognition method and system
US10262658B2 (en) * 2014-11-28 2019-04-16 Shenzhen Skyworth-Rgb Eletronic Co., Ltd. Voice recognition method and system
WO2016150001A1 (en) * 2015-03-24 2016-09-29 中兴通讯股份有限公司 Speech recognition method, device and computer storage medium
CN110415689B (en) * 2018-04-26 2022-02-15 富泰华工业(深圳)有限公司 Speech recognition device and method
US10714088B2 (en) 2018-04-26 2020-07-14 Hon Hai Precision Industry Co., Ltd. Speech recognition device and method of identifying speech
CN110415689A (en) * 2018-04-26 2019-11-05 富泰华工业(深圳)有限公司 Speech recognition equipment and method
TWI662542B (en) * 2018-04-26 2019-06-11 鴻海精密工業股份有限公司 Voice recognition device and method

Also Published As

Publication number Publication date
US7587318B2 (en) 2009-09-08
US20040117191A1 (en) 2004-06-17

Similar Documents

Publication Publication Date Title
US7587318B2 (en) Correlating video images of lip movements with audio signals to improve speech recognition
US9280539B2 (en) System and method for translating speech, and non-transitory computer readable medium thereof
US9484017B2 (en) Speech translation apparatus, speech translation method, and non-transitory computer readable medium thereof
JP7348288B2 (en) Voice interaction methods, devices, and systems
MX2008016354A (en) Detecting an answering machine using speech recognition.
US20100178956A1 (en) Method and apparatus for mobile voice recognition training
JP2014191029A (en) Voice recognition system and method for controlling voice recognition system
WO2001008384A1 (en) Cellular phone
JP7330066B2 (en) Speech recognition device, speech recognition method and its program
JP5251588B2 (en) Mobile phone terminal device and method for determining call transmission
JP6934831B2 (en) Dialogue device and program
JP3838159B2 (en) Speech recognition dialogue apparatus and program
JP2019215449A (en) Conversation auxiliary apparatus, conversation auxiliary method, and program
JP2010128766A (en) Information processor, information processing method, program and recording medium
JP2006251061A (en) Voice dialog apparatus and voice dialog method
JP2013077925A (en) Electronic apparatus
JP3891023B2 (en) Interpreter system and program
JP2003316384A (en) Real time character correction device, method, program, and recording medium for the same
JP2014092627A (en) Voice recognition device, voice recognition method and program for the same
JP2001236091A (en) Method and device for error correcting voice recognition result
JP2008051950A (en) Information processing apparatus
JP2000122678A (en) Controller for speech recogniging equipment
JPH0242494A (en) Voice input device
JP2007272123A (en) Voice operation system
JP3278595B2 (en) mobile phone

Legal Events

Date Code Title Description
AS Assignment

Owner name: BROADCOM CORPORATION, CALIFORNIA

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:SESHADRI, NAMBI;REEL/FRAME:027869/0209

Effective date: 20040112

STCB Information on status: application discontinuation

Free format text: ABANDONED -- FAILURE TO PAY ISSUE FEE

AS Assignment

Owner name: BANK OF AMERICA, N.A., AS COLLATERAL AGENT, NORTH CAROLINA

Free format text: PATENT SECURITY AGREEMENT;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:037806/0001

Effective date: 20160201

Owner name: BANK OF AMERICA, N.A., AS COLLATERAL AGENT, NORTH

Free format text: PATENT SECURITY AGREEMENT;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:037806/0001

Effective date: 20160201

AS Assignment

Owner name: AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD., SINGAPORE

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:041706/0001

Effective date: 20170120

Owner name: AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:041706/0001

Effective date: 20170120

AS Assignment

Owner name: BROADCOM CORPORATION, CALIFORNIA

Free format text: TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENTS;ASSIGNOR:BANK OF AMERICA, N.A., AS COLLATERAL AGENT;REEL/FRAME:041712/0001

Effective date: 20170119