USING A PHYSICAL PHENOMENON DETECTOR TO CONTROL OPERATION OF A SPEECH RECOGNITION ENGINE

• US 20090182560A1
• Filed: 01/16/2009
• Published: 07/16/2009
• Est. Priority Date: 01/16/2008
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: YAP INC. (Amazon.com, Inc.)
• Inventors: WHITE, Marc
• Granted Patent: US 8,326,636 B2
• Time in Patent Office: Days
• US Class Current: 704/235
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 21/00 Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility G10L19/00 takes precedence
• CPC Class Codes: H04M 1/7243 with interactive means for internal management of messages
• CPC Class Codes: H04M 1/72433 for voice messaging, e.g. dictaphones for answering incoming calls H04M1/64
• CPC Class Codes: H04M 1/72436 for text messaging, e.g. short messaging services [SMS] or e-mails
• CPC Class Codes: H04M 1/72445 for supporting Internet browser applications
• CPC Class Codes: H04M 2250/12 including a sensor for measuring a physical value, e.g. temperature or motion telephones including GPS signal receivers H04M2250/10
• CPC Class Codes: H04M 2250/74 with voice recognition means voice activated dialling H04M1/271; voice recognition algorithms G10L15/00

Abstract

A transmission device such as a cell phone or other mobile communication device includes a physical phenomenon detection device to perform a “push to talk” function by detecting the occurrence of a particular physical phenomenon and using such detection to start and stop recording an utterance for subsequent analysis by a speech recognition engine. A method of controlling operation of a speech recognition engine in response to detection of a physical phenomenon includes detecting or sensing, via a physical phenomenon detection unit, a predetermined physical phenomenon representative of an intent to invoke operation of a speech recognition engine. In response to the detection or sensing of the predetermined physical phenomenon, a signal is transmitted to a control unit in a communication device. In response to the receipt of the transmitted signal, the utterance received from a user via the communication device is recorded, and the recorded utterance is provided to a speech recognition engine for operation thereon. The user may thus effectuate operation of the speech recognition engine upon the utterance by causing the physical phenomenon to occur.

Claims

1. A method of controlling operation of a speech recognition engine in response to detection of a physical phenomenon, comprising:

detecting or sensing, via a physical phenomenon detection unit, a predetermined physical phenomenon representative of an intent to invoke operation of a speech recognition engine;

in response to the detection or sensing of the predetermined physical phenomenon, transmitting a signal to a control unit in a communication device;

in response to the receipt of the transmitted signal, recording an utterance received from a user via the communication device; and

providing the recorded utterance to a speech recognition engine for operation thereon;

wherein the user may effectuate operation of the speech recognition engine upon the utterance by causing the physical phenomenon to occur.

2. The method of claim 1, wherein the detecting or sensing step includes detecting or sensing the predetermined physical phenomenon without use of key elements.

3. The method of claim 2, wherein providing the recorded utterance to the speech recognition engine for operation thereon includes providing the recorded utterance to the speech recognition engine for recognition as a voice command.

4. The method of claim 2, wherein providing the recorded utterance to the speech recognition engine for operation thereon includes providing the recorded utterance to the speech recognition engine for transcription into text.

5. The method of claim 4, further comprising:

upon receiving transcribed text from the speech recognition engine, presenting the transcribed text to the user.

6. The method of claim 2, wherein the physical phenomenon detection unit is included in the communication device.

7. The method of claim 2, wherein the physical phenomenon detection unit is not included in the communication device.

8. The method of claim 2, wherein the physical phenomenon is a first physical phenomenon, and wherein the method further comprises:

detecting or sensing, via the physical phenomenon detection unit, a predetermined second physical phenomenon representative of an intent to further control the operation of a speech recognition engine; and

in response to the detection or sensing of the predetermined second physical phenomenon, terminating the step of recording the utterance received from the user via the communication device.

9. The method of claim 8, wherein the utterance includes an activated portion recorded after the first physical phenomenon is detected or sensed but before the second physical phenomenon is detected or sensed, and wherein operation of the speech recognition engine on the utterance is limited to the activated portion of the utterance.

10. The method of claim 9, wherein the first physical phenomenon is a first occurrence of the predetermined physical phenomenon and the second physical phenomenon is a second occurrence of the same predetermined physical phenomenon.

11. The method of claim 9, wherein the second physical phenomenon is different from the first physical phenomenon.

12. The method of claim 9, wherein detecting or sensing the first physical phenomenon is carried out via a first proximity sensor and detecting or sensing the second physical phenomenon is carried out via a second proximity sensor.

13. The method of claim 2, wherein detecting or sensing includes detecting or sensing the proximity of an object.

14. The method of claim 13, wherein detecting or sensing the proximity of the object includes detecting or sensing the proximity of an element of the communication device.

15. The method of claim 13, wherein detecting or sensing the proximity of the object includes detecting or sensing the proximity of a human body part.

16. The method of claim 15, wherein detecting or sensing the proximity of the human body part includes detecting or sensing heat radiated from a human face.

17. The method of claim 2, wherein detecting or sensing includes detecting or sensing the detection unit being tilted relative to a first axis.

18. The method of claim 17, wherein detecting or sensing further includes detecting or sensing the detection unit being tilted relative to a second axis.

19. The method of claim 2, further comprising a step of, prior to detecting or sensing, via the predetermined physical phenomenon representative of an intent to invoke operation of a speech recognition engine, receiving an input representative of an intent to place the communication device in a standby state such that subsequent detecting or sensing of the predetermined physical phenomenon is enabled.

20. The method of claim 19, wherein receiving an input includes receiving the input via one or more key element provided in the communication device.

21. A method of using a physical phenomenon detection unit in a communication device to control operation of a speech recognition engine, comprising:

manipulating a communication device so as to cause a first predetermined physical phenomenon to occur, thereby activating one or more elements of the communication device for recordation of an utterance;

speaking an utterance for receipt by the communication device;

manipulating a communication device so as to cause a second predetermined physical phenomenon to occur, thereby deactivating the one or more elements of the communication device, wherein the utterance includes an activated portion received by the communication device while the one or more elements are activated for recording; and

causing the activated portion of the utterance to be operated upon by a speech recognition engine.