Always-on keyword detector
Provided herein is an integrated circuit including, in some embodiments, a special-purpose host processor, a neuromorphic co-processor, and a communications interface between the host processor and the co-processor configured to transmit information therebetween. The special-purpose host processor is operable as a stand-alone host processor. The neuromorphic co-processor includes an artificial neural network. The co-processor is configured to enhance special-purpose processing of the host processor through the artificial neural network. In such embodiments, the host processor is a keyword identifier processor configured to transmit one or more detected words to the co-processor over the communications interface. The co-processor is configured to transmit recognized words, or other sounds, to the host processor.
1. An integrated circuit for detecting keywords, comprising:
a keyword identifier host processor operable as a stand-alone host processor and configured to identify one or more words within a received audio stream;
a neuromorphic co-processor including an artificial neural network that is configured to identify one or more desired keywords among the one or more words received from the host processor; and
a communications interface between the host processor and the co-processor configured to transmit information therebetween.
2. The integrated circuit of claim 1 , wherein the neuromorphic co-processor includes a database comprised of known keywords whereby the one or more desired keywords may be identified among the one or more words.
3. The integrated circuit of claim 2 , wherein new keywords may be added to the database and distinguished from existing keywords.
4. The integrated circuit of claim 3 , wherein the integrated circuit is configured to operate on battery power.
5. A method for detecting keywords within an audio stream, comprising:
receiving the audio stream in the form of electrical signals to a keyword identifier host processor;
transmitting the electrical signals from the host processor to a neuromorphic co-processor;
identifying one or more desired keywords within the electrical signals;
transmitting the one or more desired keywords from the neuromorphic co-processor to the host processor; and
sending one or more output signals from the host processor upon receiving the one or more desired keywords from the neuromorphic co-processor.
6. The method of claim 5 , wherein the neuromorphic co-processor includes a database comprised of known keywords whereby the one or more desired keywords may be identified within the electrical signals.
7. The method of claim 5 , wherein the one or more keywords are comprised of predefined acoustic signal other than speech.
8. An integrated circuit, comprising:
a special-purpose host processor operable as a stand-alone host processor;
a neuromorphic co-processor including an artificial neural network that is configured to enhance special-purpose processing of the host processor; and
a communications interface between the host processor and the co-processor configured to transmit information therebetween.
9. The integrated circuit of claim 8 , wherein the host processor is a keyword identifier processor configured to transmit an audio stream in the form of electrical signals to the co-processor over the communications interface, and wherein the co-processor is configured to enhance the special-purpose processing of the host processor by providing one or more detected keywords to the host processor over the communications interface.
10. The integrated circuit of claim 9 , wherein the communications interface between the host processor and the co-processor is a serial peripheral interface (“SPI”) bus or inter-integrated circuit (“I 2 C”) bus.
11. The integrated circuit of claim 10 , wherein the artificial neural network is disposed in an analog multiplier array of a number of two-quadrant multipliers in a memory sector of the integrated circuit.
12. The integrated circuit of claim 11 , wherein synaptic weights of the artificial neural network are stored in firmware of the integrated circuit, and wherein the firmware is configured for cloud-based upgrades to update the synaptic weights of the artificial neural network.
13. The integrated circuit of claim 12 , wherein the integrated circuit is configured to operate on battery power.
14. An integrated circuit for detecting keywords, comprising:
a special-purpose host processor configured to identify one or more sounds within a received audio stream and transmit the one or more sounds to a neuromorphic co-processor;
an artificial neural network comprising the neuromorphic co-processor that is configured to identify one or more desired sounds among the one or more sounds received from the host processor; and
a communications interface between the host processor and the co-processor configured to transmit information therebetween.
15. The integrated circuit of claim 14 , wherein the integrated circuit is configured to operate on battery power.
16. The integrated circuit of claim 15 , wherein the integrated circuit is configured to remain in a low-power, always-on state whereby the integrated circuit remains continuously ready to receive the audio stream.
17. The integrated circuit of claim 14 , wherein the host processor is configured to output one or more signal upon detection of the one or more desired sounds.
18. The integrated circuit of claim 14 , wherein the integrated circuit is configured for implementation within a mobile device.
19. The integrated circuit of claim 18 , wherein the integrated circuit is configured to be powered by current leakage from a battery comprising the mobile device.
20. The integrated circuit of claim 19 , wherein the integrated circuit is configured to remain in a fully aware state whereby the integrated circuit remains continuously ready to receive the audio stream.