SYSTEM AND METHOD FOR IMPROVING AN INTERACTIVE EXPERIENCE WITH A SPEECH-ENABLED SYSTEM THROUGH THE USE OF ARTIFICIALLY GENERATED WHITE NOISE

• US 20080147394A1
• Filed: 12/18/2006
• Published: 06/19/2008
• Est. Priority Date: 12/18/2006
• Status: Abandoned Application

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: DAMES, DWAYNE, METZ, BRENT D.
• Application Number:
  US11/612,170
• Publication Number:
  US 20080147394A1
• Time in Patent Office: Days
• US Class Current: 704/233
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence
• CPC Class Codes: G10L 21/02 Speech enhancement, e.g. noise reduction or echo cancellation reducing echo effects in line transmission systems H04B3/20; echo suppression in hands-free telephones H04M9/08
• CPC Class Codes: G10L 21/0208 Noise filtering

Abstract

A speech processing system for improving a user'"'"'s experience with a speech-enabled system using artificially generated white noise. The system can include an audible environment that includes at least one microphone and at least one speaker, a white noise generator, a white noise removal engine, and a speech processing system. The white noise generator can be configured to generate white noise to be audibly presented in the audible environment. This white noise can be captured in speech input and the white noise removal engine can digitally preprocess the input to remove the white noise components. The preprocessed input can be processed by the speech processing system and the speech processing system can create speech output based on the received input.

Claims

1. A speech processing system comprising:

an audible environment including at least one microphone for receiving speech input and at least one speaker for audibly presenting speech output;

a white noise generator configured to generate white noise that is audibly presented in the audible environment;

a white noise removal engine configured to digitally preprocess speech input captured by the microphone and to remove the white noise components included in the captured input; and

a speech processing system for processing the speech input after being preprocessed by the white noise removal engine and for creating the speech output.

2. The speech processing system of claim 1, wherein the white noise removal engine receives input of a signal generated by the white noise generator, wherein the received signal is subtracted from the speech input to remove the white noise components.

3. The speech processing system of claim 2, wherein the white noise removal engine is configured to perform at least one transformation to account for audible changes between white noise contributions received by the microphone and the white noise of the received signal.

4. The speech processing system of claim 1, wherein the volume level of the white noise presented in the audible environment is configurable.

5. The speech processing system of claim 4, wherein the white noise is audibly presented at an approximately constant volume.

6. The speech processing system of claim 5, wherein the configurable volume level of the white noise establishes a volume floor for the speech processing system.

7. The speech processing system of claim 4, wherein the volume level of the white noise is controllable by the speech processing system.

8. The speech processing system of claim 4, wherein a different speaker is used to audibly present the speech output than a speaker that is used to audibly present the white noise, and wherein a volume level of the speech output is programmatically linked to the volume level of the white noise.

9. The speech processing system of claim 1, wherein the white noise generator, the white noise removal engine, and the speech processing system reside within a same computer device, wherein the speaker and the microphone are communicatively linked to the computing device.

14. The method of claim 10, wherein said steps of claim 1 are performed by at least one machine in accordance with at least one computer program having a plurality of code sections that are executable by the at least one machine.

10. A method for using artificially generated white noise to raise a noise floor of a speech processing system comprising:

audibly presenting artificially generated noise at a configurable volume level to establish a noise floor for an acoustic environment;

receiving audible input containing speech obtained from the acoustic environment;

digitally processing the input containing speech to remove the artificially generated noise from the input; and

audibly presenting output containing artificially generated speech to the acoustic environment, wherein the artificially generated speech is generated by a speech processing system, and wherein the speech processing system receives the processed input.

11. The method of claim 10, wherein the presented artificially generated noise is presented at an approximately constant volume level.

12. The method of claim 10, further comprising:

sampling input from the acoustic environment to determine an ambient noise level;

automatically calculating a desired noise floor based upon results of the sampling step; and

automatically adjusting the configurable volume level to achieve the desired noise floor.

13. The method of claim 10, further comprising:

a noise removal engine receiving a signal from a noise generator, which generates the artificially generated noise, said signal including a waveform of the artificially generated noise; and

digitally subtracting the waveform of the artificially generated noise from the received audible input.

15. The method of claim 10, wherein the steps of claim 10 are performed by at least one of a service agent and a computing device manipulated by the service agents, the steps being performed in response to a service request.

16. A method for improving a user'"'"'s experience with a speech-enabled system using artificially generated white noise comprising:

producing white noise in an acoustic environment at an established volume level;

audibly presenting automatically generated speech output in the acoustic output;

capturing speech input from the acoustic environment;

removing the white noise from the captured input to generate clean speech input; and

speech-to-text converting the clean speech input.

17. The method of claim 16, further comprising:

changing the established volume level at which the white noise is produced; and

automatically adjusting a volume level of the automatically generated speech output in accordance with the changed volume level of the white noise.

18. The method of claim 16, wherein the established volume level is a configurable value and is an approximately constant volume level.

19. The method of claim 16, wherein the speech-to-text converting step is performed by a speech processing system that also generates the speech output, said speech processing system being configured to establish the volume level of the produced white noise.

20. The method of claim 16, wherein said steps of claim 16 are performed by at least one machine in accordance with at least one computer program having a plurality of code sections that are executable by the at least one machine.