Acoustic speech recognizer system and method

Details Acoustic speech recognizer system and method Acoustic speech recognizer system and method

1999-01-13

2003-06-03

Knepper, David D. (Department: 2654)

Data processing: speech signal processing, linguistics, language

Speech signal processing

Application

C704S253000, C704S260000, C704S233000, C379S080000, C379S088010

Reexamination Certificate

active

06574601

ABSTRACT:

BACKGROUND OF THE INVENTION
The present invention relates to speech recognition systems, and, more particularly, to an acoustic speech recognizer system and method.
Speech recognition systems are known which allow vocal inputs to supplement or supplant other methods for inputting data and information, for example, to computer systems. One such system is the Bell Labs Acoustic Speech Recognizer (BLASR), available from LUCENT TECHNOLOGIES, INC., which may be used to implement an Internet and/or World Wide Web browser responsive to vocal commands, as described in commonly-assigned, U.S. patent application Ser. No. 09/168,405 of Michael Brown et al., entitled WEB-BASED PLATFORM FOR INTERACTIVE VOICE RESPONSE (IVR), filed Oct. 6, 1998, which is incorporated herein by reference.
However, speech recognition systems with barge-in capabilities mix different speech during barge-in, which badgers a speech recognition server with meaningless speech packets, and so increases the processing load of the client.
SUMMARY OF THE INVENTION
An acoustic speech recognizer system integrates a barge-in detector with an adaptive speech endpoint detector for detecting endpoints; that is, the initiation and termination of speech, to permit barge-in regardless of the intensity of conflicting output speech, by using continuously adapted barge-in thresholds. Advantageously, badgering of the speech processors is avoided. The adaptive speech endpointer detector is used in speech recognition applications, such as telephone-based Internet browsers, to determine barge-in events during the processing of speech. Continuous operation may also be performed by the adaptive speech endpoint detector to implement a voice activated web browser without the need for extraneous commands such as a push-to-talk command.
More specifically, the endpointer system includes a signal energy level estimator for estimating signal levels in speech data; a noise energy level estimator for estimating noise levels in the speech data; and a barge-in detector for increasing a threshold used in comparing the signal levels and the noise levels to detect the barge-in event in the speech data corresponding to a speech prompt during speech recognition.


REFERENCES:
patent: 5708704 (1998-01-01), Fisher
patent: 5765130 (1998-06-01), Nguyen
patent: 5937379 (1999-08-01), Takagi
patent: 5956675 (1999-09-01), Setlur et al.
patent: 5978763 (1999-11-01), Bridges
patent: 5991726 (1999-11-01), Immarco et al.
patent: 6061651 (2000-05-01), Nguyen
patent: 6144938 (2000-11-01), Surace et al.
patent: 6173266 (2001-01-01), Marx et al.
patent: 6195417 (2001-02-01), Dans
patent: 6408272 (2002-06-01), White et al.
patent: 80/00757 (1980-04-01), None
M. Padmanabhan et al.; Speech recognition performance on a voicemaill transcription task; 1998, IEEE, pp. 913-916.*
VoiveXML Version 1.0; Boyer et al.; Voice XML Forum technical working group; Mar. 2000.*
J.F. Lynch, Jr. et al.; “Speech/Silence Segmentation for Realtime Coding Via Rule Based Adaptive Endpoint Detection”, IEEE International Conference on Acoustics, Speech, and Signal Processing (1987), pp. 1348-1351.
U.S. Appln. No. 09/168,405 of Michael Brownet al., filed Oct. 6, 1998, entitled “Web-Based Platform for Interactive Voice Response (IVR)”.

Affiliated with

Also associated with

Rating

Acoustic speech recognizer system and method does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Acoustic speech recognizer system and method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Acoustic speech recognizer system and method will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3132700