Buffered reprocessing for multi-microphone automatic speech recognition assist
View Patent ↗Systems and methods for assisting automatic speech recognition (ASR) are provided. An example system includes a buffer operable to store sensor data. The sensor data includes an acoustic signal, the acoustic signal representing at least one captured sound. The system includes a processor communicatively coupled to the buffer and being operable to store received sensor data in the buffer. The received sensor data is analyzed to produce new parameters associated with the sensor data. The buffered sensor data is processed based at least on the new parameters. The processing may include separating clean voice from noise in the acoustic signal. The processor is further operable to provide at least the processed sensor data (for example, the clean voice) to an ASR system operable to receive and process the processed sensor data at a speed faster than real time. The new parameters may also be provided to the ASR system.
1. A system for assisting automatic speech recognition (ASR), the system comprising:
a buffer, operable to receive sensor data, the sensor data including at least one acoustic signal, the at least one acoustic signal representing at least one captured sound; and
a processor, communicatively coupled to the buffer, the processor being operable to:
store the received sensor data in the buffer;
analyze the received sensor data to produce new parameters associated with the sensor data, wherein the analyzing includes determining a direction of arrival of the at least one acoustic signal;
process the stored sensor data based at least in part on the new parameters; and
provide at least the processed sensor data to an ASR system.
2. The system of claim 1 , wherein the ASR system is operable to receive and process the processed sensor data at a speed faster than real time.
3. The system of claim 1 , wherein the processor is further operable to replace the sensor data in the buffer with Phell a next portion of the sensor data.
4. The system of claim 1 , wherein:
the processor is further operable to provide the new parameters to the ASR system; and
the ASR system is configured to analyze the processed sensor data based, at least in part, on the new parameters.
5. The system of claim 1 , wherein:
the processor is further operable to provide the new parameters to the ASR system; and
in response to receiving a notification from the ASR system, the processor is further operable to send the sensor data to the ASR system for further processing based, at least in part, on the new parameters.
6. The system of claim 1 , wherein:
the ASR system is located remotely; and
the processor is communicatively coupled to the ASR system via a high speed network.
7. The system of claim 1 , wherein the new parameters include one or more of the following: inter-microphone energy level differences, inter-microphone phase differences, acoustic signal energy, estimated pitch, and estimated saliency of the pitch.
8. The system of claim 1 , wherein processing the stored sensor data includes separating a clean voice from a noise in the at least one acoustic signal, the processed sensor data provided to the ASR system including the clean voice.
9. The system of claim 8 , wherein the separating includes performing at least one of the following: noise suppression and noise reduction.
10. The system of claim 1 , wherein the sensor data are provided by one or more of the following:
a sound sensor configured to capture the at least one acoustic signal;
a motion sensor;
an environment sensor;
a radio sensor; and
a light sensor.
11. A non-transitory computer-readable storage medium having embodied thereon instructions, which, when executed by one or more processors, perform steps of a method, the method comprising:
receiving sensor data, the sensor data including at least one acoustic signal, the at least one acoustic signal representing at least one captured sound;
storing the received sensor data in a buffer;
analyzing the received sensor data to produce new parameters;
processing the stored sensor data based at least in part on the new parameters; and
providing the processed sensor data to an automatic speech recognition (ASR) system, the ASR system being operable to receive and process the processed sensor data at a speed faster than real time.
12. A method for assisting automatic speech recognition (ASR), the method comprising:
receiving sensor data, the sensor data including at least one acoustic signal, the at least one acoustic signal representing at least one captured sound;
storing the received sensor data in a buffer;
analyzing the received sensor data, including determining a direction of arrival of the at least one acoustic signal, to produce new parameters;
processing the stored sensor data based at least in part on the new parameters; and
providing the processed sensor data to an ASR system.
13. The method of claim 12 , wherein the ASR system, to which the processed sensor data is provided, is operable to receive and process the processed sensor data at a speed faster than real time.
14. The method of claim 12 , further comprising replacing the sensor data in the buffer with a next portion of the sensor data.
15. The method of claim 12 , further comprising:
providing the new parameters to the ASR system, wherein the ASR system is configured to analyze the processed sensor data based at least in part on the new parameters.
16. The method of claim 12 , wherein the ASR system is located remotely and operable to receive the processed sensor data via a high-speed network.
17. The method of claim 12 , wherein the new parameters include one or more of the following: inter-microphone energy level differences, inter-microphone phase differences, acoustic signal energy, estimated pitch, and estimated saliency of the pitch.
18. The method of claim 12 , wherein processing the stored sensor data includes separating a clean voice from a noise in the at least one acoustic signal, the processed sensor data provided to the ASR system including the clean voice.