Speech interface device with caching component
A speech interface device is configured to receive response data from a remote speech processing system for responding to user speech. This response data may be enhanced with information such as a remote ASR result(s) and a remote NLU result(s). The response data from the remote speech processing system may include one or more cacheable status indicators associated with the NLU result(s) and/or remote directive data, which indicate whether the remote NLU result(s) and/or the remote directive data are individually cacheable. A caching component of the speech interface device allows for caching at least some of this cacheable remote speech processing information, and using the cached information locally on the speech interface device when responding to user speech in the future. This allows for responding to user speech, even when the speech interface device is unable to communicate with a remote speech processing system over a wide area network.
1. A method comprising:
generating, by a device, audio data that represents user speech;
performing, by the device, speech processing using the audio data to generate a first result;
determining, by the device that a remote system is unavailable for processing the first result;
determining, by the device and while the remote system is unavailable, that the first result matches with a second result previously stored in a memory of the first device;
determining, by the device, that command information associated with the second result is stored in the memory, the command information being previously received from the remote system and being based on previous interactions between the device and the remote system; and
performing, by the device, an action specified in the command information, the action being associated with the second result.
2. The method of claim 1 , further comprising:
generating, by the device, second audio data that represents second user speech;
performing the speech processing using the second audio data to determine the action;
retrieving the second result from the memory of the device, and
performing, by the device and using the second result, the action.
3. The method of claim 1 , wherein performing the speech processing comprises performing automatic speech recognition (ASR) processing on the audio data to generate text data.
4. The method of claim 1 , wherein the second result includes a model trained by one or more remote components, and wherein performing the speech processing comprises performing the speech processing on the audio data using the model.
5. The method of claim 1 , wherein the second result is used to identify a second device collocated in an environment with the device, and further comprising:
sending a command to the second device, the command instructing an action to be performed at the second device.
6. The method of claim 1 , wherein each of the previous interactions comprises an interaction where a previous user speech is spoken in an environment of the device and a microphone of the device generates respective audio data representing the previous user speech.
7. The method of claim 1 , wherein the second result is associated with a previous user speech that was detected by a second device.
8. The method of claim 7 , wherein the second device is collocated in an environment with the device.
9. A device comprising:
one or more processors; and
memory storing computer-executable instructions that, when executed by the one or more processors, cause the device to:
generate audio data that represents user speech;
perform speech processing using the audio data to generate a first result;
determine that a remote system is unavailable for processing the first result;
determine that the first result matches a second result stored in a memory of the device;
determine that command information associated with the second result is stored in the memory, the command information being previously received from the remote system and being based on previous interactions between the device and the remote system; and
perform an action specified in the command information, the action being associated with the second result.
10. The device of claim 9 , wherein the instructions, when executed by the one or more processors, further cause the device to:
generate second audio data that represents second user speech;
perform the speech processing using the second audio data to determine the action;
retrieve the second result from the memory of the device, and
perform, using the second result, the action.
11. The device of claim 9 , wherein performing the speech processing comprises performing automatic speech recognition (ASR) processing on the audio data to generate text data.
12. The device of claim 9 , wherein the second result includes a model trained by one or more remote components, and wherein performing the speech processing comprises performing the speech processing on the audio data using the model.
13. The device of claim 9 , wherein the second result is used to identify a second device collocated in an environment with the device, and wherein the instructions, when executed by the one or more processors, further cause the device to:
send a command to the second device, the command instructing an action to be performed at the second device.
14. The device of claim 9 , wherein each of the previous interactions comprises an interaction where a previous user speech is spoken in an environment of the device and a microphone of the device generates respective audio data representing the previous user speech.
15. The device of claim 9 , further comprising a microphone, and wherein each of the previous interactions comprises an interaction where a user spoke in an environment of the device and the microphone generated respective audio data representing the user speech.
16. The device of claim 9 , wherein the second result is associated with a previous user speech that was detected by a second device.
17. The device of claim 16 , wherein the second device is collocated in an environment with the device.
18. The device of claim 9 , further comprising a microphone, and wherein the microphone generated the audio data that represents the user speech.