Method and apparatus to perform speech recognition over a data channel

• US 20030110042A1
• Filed: 12/07/2001
• Published: 06/12/2003
• Est. Priority Date: 12/07/2001
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Stanford, Michael
• Granted Patent: US 7,162,414 B2
• Time in Patent Office: Days
• US Class Current: 704/500
• CPC Class Codes: G10L 15/08 Speech classification or search

Abstract

A method and apparatus to perform speech recognition over a data channel is described.

Claims

1. A method to perform speech recognition, comprising:

communicating voice information encoded using a voice compression algorithm over a voice channel;

receiving a request for voice information for a speech recognition system; and

communicating said requested voice information encoded using said voice compression algorithm over a data channel.

2. The method of claim 1, wherein said data channel provides higher quality voice information than said voice channel.

3. The method of claim 1, wherein said communicating said voice information comprises:

receiving an analog audio waveform representing speech;

converting said analog audio waveform into a digital audio signal;

encoding said digital audio signal using said voice compression algorithm; and

transmitting said digital audio signal over said voice channel.

4. The method of claim 1, wherein said communicating said requested voice information comprises:

receiving an analog audio waveform representing speech;

converting said analog audio waveform into a digital audio signal;

encoding said digital audio signal using said voice compression algorithm; and

transmitting said digital audio signal over said data channel.

5. The method of claim 1, wherein said request comprises a prompt for a voice command.

6. The method of claim 4, wherein said transmitting said digital audio signal over said data channel comprises:

receiving a first signal from said speech recognition system to initiate sending said requested voice information over said-data channel; and

transmitting said digital audio signal over said data channel in accordance with said first signal.

7. The method of claim 6, further comprising:

receiving a second signal from said speech recognition system to terminate sending said requested voice information over said data channel; and

terminating transmission of said digital audio signal over said data channel in accordance with said second signal.

8. The method of claim 4, wherein said transmitting said digital audio signal over said data channel comprises:

monitoring for a signal from a user to initiate sending said requested voice information over said data channel;

receiving said signal from said user; and

transmitting said digital audio signal over said data channel while said signal is received.

9. The method of claim 1, further comprising:

sending said request for voice information for a speech recognition system over said voice channel;

receiving said requested voice information over said data channel; and

performing speech recognition using said requested voice information.

10. The method of claim 9, further comprising:

sending a first signal over said voice channel to initiate sending said requested voice information over said data channel; and

receiving said requested voice information over said data channel in accordance with said first signal.

11. The method of claim 10, further comprising sending a second signal over said voice channel to terminate sending said requested voice information over said data channel.

12. A mobile system, comprising:

a vocoder to encode speech using a voice compression algorithm;

a transceiver to communicate said encoded speech over one of a voice channel and a data channel; and

an automatic speech recognition (ASR) control module to control when said encoded speech is communicated over said voice channel and said data channel.

13. The mobile system of claim 12, further comprising an ASR input module to generate a signal from a user, said ASR control module to receive said signal and cause said encoded speech to be communicated over said data channel while said signal is received.

14. The mobile system of claim 13, wherein said ASR input module comprises:

an input device to move between a first position and a second position;

a position sensor to determine when said input device is in said first position and said second position;

a position signal generating module to generate a signal indicating when said input device is in one of said first and second positions.

15. The mobile system of claim 14, wherein said input device comprises a button.

16. The mobile system of claim 12, further comprising:

a microphone to receive analog speech signals; and

an analog-to-digital converter to convert said analog speech signals into digital speech signals.

17. A wireless communication system, comprising:

a speech recognition system;

a base station in communication with said speech recognition system; and

a mobile station to communicate with said base station using radio-frequencies, said mobile station to communicate speech encoded using a voice compression algorithm to said speech recognition system over a voice channel and a data channel.

18. The wireless communication system of claim 17, wherein said mobile station comprises:

a vocoder to encode said speech using said voice compression algorithm;

a transceiver to communicate said encoded speech over one of said voice channel and said data channel; and

an automatic speech recognition (ASR) control module to control when said encoded speech is communicated over said voice channel and said data channel.

19. The wireless communication system of claim 18, further comprising an ASR input module to generate a signal from a user, said ASR control module to receive said signal and cause said encoded speech to be communicated over said data channel while said signal is received.

20. The mobile system of claim 19, wherein said ASR input module comprises:

an input device to move between a first position and a second position;

a position sensor to determine when said input device is in said first position and said second position;

a position signal generating module to generate a signal indicating when said input device is in one of said first and second positions.

21. A method to perform speech recognition, comprising:

communicating voice information over a voice channel;

receiving a request for voice information for a speech recognition system; and

communicating said requested voice information over a data channel.

22. The method of claim 21, wherein said communicating said requested voice information comprises:

receiving an analog audio waveform representing speech;

converting said analog audio waveform into a digital audio signal; and

transmitting said digital audio signal over said data channel.

23. The method of claim 21, wherein said request comprises a prompt for a voice command.

24. The method of claim 22, wherein said transmitting said digital audio signal over said data channel comprises:

receiving a first signal from said speech recognition system to initiate sending said requested voice information over said data channel; and

transmitting said digital audio signal over said data channel in accordance with said first signal.

25. The method of claim 24, further comprising:

receiving a second signal from said speech recognition system to terminate sending said requested voice information over said data channel; and

terminating transmission of said digital audio signal over said data channel in accordance with said second signal.

26. The method of claim 22, wherein said transmitting said digital audio signal over said data channel comprises:

monitoring for a signal from a user to initiate sending said requested voice information over said data channel;

receiving said signal from said user; and

transmitting said digital audio signal over said data channel while said signal is received.