Methods and apparatus for performing speech recognition and using speech recognition results

• US 20020065657A1
• Filed: 11/30/2000
• Published: 05/30/2002
• Est. Priority Date: 11/30/2000
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Telesector Resources Group Incorporated
• Inventors: Reding, Craig L., Levas, Suzi
• Granted Patent: US 6,915,262 B2
• Time in Patent Office: Days
• US Class Current: 704/246
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 17/04 Training, enrolment or model building
• CPC Class Codes: G10L 2015/221 Announcement of recognition results
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 2207/18 wireless networks
• CPC Class Codes: H04M 3/42204 Arrangements at the exchange for service or number selection by voice at the terminal H04M1/27

Abstract

Techniques for generating, distributing, and using speech recognition models are described. A shared speech processing facility is used to support speech recognition for a wide variety of devices with limited capabilities including business computer systems, personal data assistants, etc., which are coupled to the speech processing facility via a communications channel, e.g., the Internet. Devices with audio capture capability record and transmit to the speech processing facility, via the Internet, digitized speech and receive speech processing services, e.g., speech recognition model generation and/or speech recognition services, in response. The Internet is used to return speech recognition models and/or information identifying recognized words or phrases. Thus, the speech processing facility can be used to provide speech recognition capabilities to devices without such capabilities and/or to augment a device'"'"'s speech processing capability. Voice dialing, telephone control and/or other services are provided by the speech processing facility in response to speech recognition results.

Claims

1. A speech processing method, comprising the steps of:

operating a first device to receive speech;

transmitting the received speech from the first device to a remote speech processing facility via the Internet;

performing a speech recognition operation on said speech at the remote speech processing facility; and

performing an action based on the outcome of the speech recognition operation.

2. The method of claim 1, wherein the step of performing an action includes:

transmitting information generated from said speech recognition operation via the Internet.

3. The method of claim 2,

wherein said information generated from said speech recognition operation is information indicating at least some of the recognized content of said speech.

4. The method of claim 2,

wherein said step of performing an action includes:

retrieving from memory a telephone number associated with a name which was recognized in said speech; and

transmitting the telephone number to the first device via the Internet.

5. The method of claim 4, further comprising the steps of:

operating the first device to receive the telephone number transmitted via the Internet; and

operating the first device to initiate a telephone call using the telephone number.

6. A method of operating a speech processing facility coupled to the Internet, the method comprising the steps of:

receiving speech data transmitted over the Internet;

performing a speech recognition operation using the received speech data; and

performing an action based on the results of the speech recognition operation performed using the received speech information.

7. The method of claim 6, wherein the step of performing an action includes:

initiating a telephone call using a telephone number associated with a name recognized by performing said speech recognition operation.

8. The method of claim 7, further comprising the steps of:

initiating an additional telephone call to a telephone where the human being who was the source of speech corresponding to said speech data can be contacted; and

bridging said telephone call and said additional telephone call.

9. The method of claim 8, further comprising the step of:

receiving via the Internet said telephone number where the human being who was the source of speech corresponding to said speech data can be contacted.

10. The method of claim 6, wherein said speech data includes at least one of digitized speech and speech feature vectors.

11. The method of claim 6, wherein the step of performing an action includes:

transmitting the speech data via the Internet to another speech processing facility for additional processing.

12. The method of claim 6, wherein the step of performing an action includes:

transmitting information generated from said speech recognition operation to a device via the Internet.

13. The method of claim 12, further comprising the step of:

operating said device to fill out at least a portion of a form based on the information generated from said speech recognition operation.

14. The method of claim 6, wherein the step of performing an action includes:

controlling a telephony device as a function of recognized speech.

15. The method of claim 15, wherein the telephony device is coupled to the Internet and a telephone line;

and

wherein the telephony device is controlled to perform a dialing operation as a function of recognized speech.

17. The method of claim 16, wherein the second vocabulary is larger than the first vocabulary.

18. The method of claim 17, wherein the first vocabulary includes words corresponding to a first set of names and the second vocabulary includes words corresponding to a second set of names.

19. The method of claim 15, further comprising:

when the first device recognizes a word representing a name with which a telephone number is associated, the first device;

initiating a telephone call to said telephone number.

20. The method of claim 15, the method further comprising the steps of:

performing a feature vector extraction operation on said received speech; and

including the extracted speech feature vectors in said speech data.

21. The method of claim 20, further comprising the step of:

including a digital recording of said received speech in said speech data.

22. The method of claim 15, further comprising the step of:

receiving a telephone number from said remote speech processing facility; and

initiating a telephone call by dialing the received telephone number.

23. The method of claim 22, wherein when the first device fails to recognize any words in the first vocabulary in said received speech, the first device further performs the step of:

transmitting to the remote speech processing facility a contact telephone number corresponding to a telephone at which a human being who was the source of the received speech can be contacted.

25. The system of claim 24, wherein said device further comprises:

a speech recognizer for performing a speech recognition operation on said received speech prior to transmitting it to said speech recognition facility.

26. The system of claim 25, wherein said device further includes:

means for dialing a telephone number associated with a name when said speech recognizer identifies the name in received speech.

28. The speech processing facility of claim 27, wherein the means for performing an action based on the results of the speech recognition operation include:

a dialer for dialing a telephone number associated with a recognized name.

29. The speech processing facility of claim 28, wherein the means for performing an action further includes:

call bridging circuitry for bridging a telephone call initiated by the dialer dialing a telephone call with another telephone call.

30. The speech processing facility of claim 28, further comprising:

a memory device including speech recognition models of names and telephone numbers associated with the modeled names.

16. A speech processing method, the method comprising the steps of:

operating a first device to receive speech;

operating the first device to perform a speech recognition operation on said received speech using a first vocabulary including words; and

when the first device fails to recognize any words in the first vocabulary in said received speech, performing the step of;

transmitting digital speech data via the Internet to a remote speech processing facility capable of performing a speech recognition operation using a second vocabulary which is different from said first vocabulary.

24. A speech processing system comprising, comprising:

a speech processing facility coupled to the Internet, the speech processing facility including;

i. a receiver for receiving speech data from the Internet;

ii. a speech recognizer for performing a speech recognition operation on speech received via the Internet; and

iii. means for performing an action based on the outcome of the speech recognition operation; and

a device coupled to the Internet, the device including;

i. a microphone for receiving speech; and

ii. a transmitter coupled to the Internet for transmitting speech data including at least some of the received speech in digital form to said speech processing facility.

27. A speech processing facility, comprising:

means for receiving speech data transmitted over the Internet;

a speech recognizer for performing a speech recognition operation on the received speech data; and

means for performing an action based on the results of the speech recognition operation performed using the received speech information.