Speech recognition aided by lateral profile image
AssigneeInternational Business Machines Corporation
InventorsChengjun Julian Chen, Frederick Yung-Fung Wu, James T. Yeh
Life of this patent
Estimating the term…
Abstract
An apparatus and a method for imaging the mouth area laterally to produce reliable measurements of mouth and lip shapes for use in assisting the speech recognition task. A video camera is arranged with a headset and a microphone to capture a lateral profile image of a speaker. The lateral profile image is then used to compute features such as lip separation, lip shape and intrusion depth parameters. The parameters are used in real time, during speech recognition process to characterize and discriminate spoken phonemes to produce a high degree of accuracy in automatic speech recognition processing, especially in a noisy environment.
Claims27
An apparatus for acquiring video images and video data for use with automated acoustic speech recognition systems, comprising:
- a video camera positioned to capture lateral views of a speaker's mouth area while uttering speech, the video camera capturing lateral images;
- and
- first means for computing image features from a set of only captured lateral images,
- wherein the computed image features are used with acoustics signals of the uttered speech for automated speech recognition.
A method of acquiring video image data for use with an automated acoustic speech recognition system, the method comprising:
- capturing lateral images of a speaker's face focusing in a range of speaker's lips;
- extracting speaker's profile from a set of only captured lateral images;
- and
- computing image features from the speaker's profile,
- wherein the computed image features are used with acoustics signals for automated speech recognition.
Dependencies are inferred from the claim text.