System and method for mobile automatic speech recognition
A system and method of updating automatic speech recognition parameters on a mobile device are disclosed. The method comprises storing user account-specific adaptation data associated with ASR on a computing device associated with a wireless network, generating new ASR adaptation parameters based on transmitted information from the mobile device when a communication channel between the computing device and the mobile device becomes available and transmitting the new ASR adaptation data to the mobile device when a communication channel between the computing device and the mobile device becomes available. The new ASR adaptation data on the mobile device more accurately recognizes user utterances.
1. A method comprising:
identifying, by a system including a processor, automatic speech recognition parameters for use by a speech recognition system; and
transmitting, by the system and to the speech recognition system, a set of automatic speech recognition adaptation parameters that enable updating the automatic speech recognition parameters of the speech recognition system.
2. The method of claim 1 , wherein identifying the automatic speech recognition parameters comprises:
receiving, at the system and from the speech recognition system, speech recognition data associated with speech recognition performed by the speech recognition system; and
identifying, at the system, the automatic speech recognition parameters based on the speech recognition data and user-specific information stored at the system that are associated with a user of the speech recognition system.
3. The method of claim 2 , further comprising:
generating, by the system, the set of automatic speech recognition adaptation parameters based on the automatic speech recognition parameters.
4. The method of claim 3 , wherein the generating the automatic speech recognition adaptation parameters comprises synchronizing the automatic speech recognition parameters with the user-specific information.
5. The method of claim 3 , wherein the generating the automatic speech recognition adaptation parameters comprises normalizing the automatic speech recognition parameters using frequency wrapping.
6. The method of claim 5 , wherein the frequency warping is performed by selecting a singular linear warping function using normalized automatic speech recognition parameters.
7. The method of claim 2 , wherein the user-specific information comprises information reflective of common terminologies spoken by a user associated with the speech recognition system and environments in which the speech recognition system is used.
8. The method of claim 2 , wherein the speech recognition data include auditory information indicating an environment in which the speech recognition system performed the speech recognition and multi-modal input from a user associated with the speech recognition system which is used in performing the speech recognition.
9. A network device, comprising:
a memory having computer-readable instructions stored therein; and
a processor configured to execute the computer-readable instructions to,
identify automatic speech recognition parameters for use by a speech recognition system; and
transmit, to the speech recognition system, a set of automatic speech recognition adaptation parameters that enable updating the automatic speech recognition parameters of the speech recognition system.
10. The network device of claim 9 , wherein the processor is further configured to generate the set of automatic speech recognition adaptation parameters based on the automatic speech recognition parameters.
11. The network device of claim 10 , wherein the generating of the automatic speech recognition adaptation parameters further comprises synchronizing the automatic speech recognition parameters with user-specific information.
12. The network device of claim 10 , wherein the generating of the automatic speech recognition adaptation parameters further comprises normalizing the automatic speech recognition parameters using frequency wrapping.
13. The network device of claim 9 , wherein the speech recognition system is a mobile device.
14. The network device of claim 13 , wherein the network device is a server that is communicatively coupled to the speech recognition system.
15. The network device of claim 14 , wherein the mobile device and the server communicate intermittently.
16. A non-transitory computer-readable medium having computer-readable instructions stored thereon, which when executed by one or more processors, cause the one or more processors at a network node to perform operations comprising:
identifying automatic speech recognition parameters for use by a speech recognition system; and
transmitting, to the speech recognition system, a set of automatic speech recognition adaptation parameters that enable updating the automatic speech recognition parameters of the speech recognition system.
17. The non-transitory computer-readable medium of claim 16 , wherein identifying the automatic speech recognition parameters comprises:
receiving, from the speech recognition system, speech recognition data associated with speech recognition performed by the speech recognition system; and
identifying the automatic speech recognition parameters based on the speech recognition data and stored user-specific information.
18. The non-transitory computer-readable medium of claim 17 , wherein the speech recognition data include auditory information indicating an environment in which the speech recognition system performed the speech recognition and multi-modal input from a user associated with the speech recognition system which is used in performing the speech recognition.
19. The non-transitory computer-readable medium of claim 16 , wherein the operations further comprise:
generating the set of automatic speech recognition adaptation parameters based on the automatic speech recognition parameters.
20. The non-transitory computer-readable medium of claim 19 , wherein the generating of the automatic speech recognition adaptation parameters further comprises normalizing the automatic speech recognition parameters using frequency wrapping.