Hosted voice recognition system for wireless devices

• US 9,009,055 B1
• Filed: 04/29/2013
• Issued: 04/14/2015
• Est. Priority Date: 04/05/2006
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Canyon IP Holdings LLC (Intellectual Ventures LLC)
• Inventors: Jablokov, Victor R., Jablokov, Igor R., White, Marc
• Primary Examiner(s): Godbold, Douglas
• Application Number:
  US13/872,928
• Time in Patent Office: 715 Days
• Field of Search: 704/230, 704/231, 704/270
• US Class Current: 704/270
• CPC Class Codes: G06Q 30/0251 Targeted advertisements
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• 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: H04L 51/066 Format adaptation, e.g. format conversion or compression
• CPC Class Codes: H04L 51/58 Message adaptation for wireless communication

Abstract

Methods, systems, and software for converting the audio input of a user of a hand-held client device or mobile phone into a textual representation by means of a backend server accessed by the device through a communications network. The text is then inserted into or used by an application of the client device to send a text message, instant message, email, or to insert a request into a web-based application or service. In one embodiment, the method includes the steps of initializing or launching the application on the device; recording and transmitting the recorded audio message from the client device to the backend server through a client-server communication protocol; converting the transmitted audio message into the textual representation in the backend server; and sending the converted text message back to the client device or forwarding it on to an alternate destination directly from the server.

Claims

1. A computer-implemented method comprising:

receiving, at a device, an audio input;

transmitting, to a server, data representing a first portion of the audio input, as the first portion is received at the device;

receiving, from the server, an identifier associated with the first portion of the audio input;

transmitting, to the server, the identifier and a request for partial speech recognition results determined from the first portion of the audio input; and

receiving, from the server, the partial speech recognition results determined from the first portion of the audio input, while a second portion of the audio input is being received at the device.

2. The computer-implemented method of claim 1, further comprising presenting the partial speech recognition results at the device, while the second portion of the audio input is being received at the device.

3. The computer-implemented method of claim 2, further comprising:

transmitting, to the server, data representing the second portion of the audio input, as the second portion is received at the device;

receiving, from the server, second partial speech recognition results determined from at least the second portion of the audio input; and

presenting the second partial speech recognition results at the device.

4. The computer-implemented method of claim 2, wherein the partial speech recognition results are presented using a text-to-speech engine.

5. The computer-implemented method of claim 1, wherein the device is configured to begin receiving the audio input when a button is pressed, and finish receiving the audio input when the button is released.

6. The computer-implemented method of claim 1, wherein the transmitting of the data representing the first portion to the server is performed by a first thread and the receiving of the partial speech recognition results from the server is performed by a second thread.

7. The computer-implemented method of claim 1, wherein the audio input comprises a plurality of spoken words and the partial speech recognition results comprise at least one word represented as text.

8. The computer-implemented method of claim 1, wherein the message is edited at the device before it is transmitted.

9. A non-transitory computer storage medium storing instructions that, when executed, cause a processor to:

receive an audio input;

transmit, to a server, data representing a first portion of the audio input, as the first portion is received;

receive an identifier associated with the first portion of the audio input from the server;

request the partial speech recognition results from the server using the identifier; and

receive, from the server, the partial speech recognition results determined from the first portion of the audio input, while receiving a second portion of the audio input.

10. The non-transitory computer storage medium of claim 9, further comprising instructions that, when executed, cause a processor to present the partial speech recognition results while the second portion of the audio input is being received.

11. The non-transitory computer storage medium of claim 10, further comprising instructions that, when executed, cause a processor to:

transmit, to the server, data representing the second portion of the audio input, as the second portion is received;

receive, from the server, second partial speech recognition results determined from at least the second portion of the audio input; and

present the second partial speech recognition results.

12. The non-transitory computer storage medium of claim 10, wherein the partial speech recognition results are presented using a text-to-speech engine.

13. The non-transitory computer storage medium of claim 9, further comprising instructions that, when executed, cause a processor to begin receiving the audio input when a button is pressed, and finish receiving the audio input when the button is released.

14. The non-transitory computer storage of claim 9, wherein the audio input comprises a plurality of spoken words and the partial speech recognition results comprise at least one word represented as text.

15. The non-transitory computer storage medium of claim 9, wherein the transmitting of data representing the first portion to the server is performed by a first thread and the receiving of the partial speech recognition results from the server is performed by a second thread.

16. The non-transitory computer storage medium of claim 9, further comprising instructions that, when executed, cause a processor to receive commands for editing the message before it is transmitted.

17. A mobile apparatus comprising:

memory configured to store audio input; and

a processor in communication with the memory, the processor configured to;

receive an audio input;

transmit, to a server, data representing a first portion of the audio input, as the first portion is received;

receive an identifier associated with the first portion of the audio input from the server;

request the partial speech recognition results from the server using the identifier; and

receive, from the server, the partial speech recognition results determined from the first portion of the audio input, while receiving a second portion of the audio input.

18. The mobile apparatus of claim 17, wherein the processor is further configured to present the partial speech recognition results while the second portion of the audio input is being received.

19. The mobile apparatus of claim 18, wherein the processor is further configured to:

transmit, to the server, data representing the second portion of the audio input, as the second portion is received;

receive, from the server, second partial speech recognition results determined from at least the second portion of the audio input;

and present the partial speech recognition results.

20. The mobile apparatus of claim 18, wherein the partial speech recognition results are presented using a text-to-speech engine.

21. The mobile apparatus of claim 17, wherein the processor is further configured to begin receiving the audio input when a button is pressed, and finish receiving the audio input when the button is released.

22. The mobile apparatus of claim 17, wherein the processor is further configured to transmit the data representing the first portion to the server using a first thread and receive the partial speech recognition results from the server using a second thread.

23. The mobile apparatus of claim 17, wherein the audio input comprises a plurality of spoken words and the partial speech recognition results comprise at least one word represented as text.

24. The mobile apparatus of claim 17, wherein the processor is further configured to receive commands for editing the message before it is transmitted.