Speech recognition systems and methods for concurrent voice commands
Systems and methods are provided that support concurrent voice commands from different users. A speech recognition system includes a first input interface to receive a first audio input, a second input interface to receive a second audio input, an output interface, a first recognition module coupled to the first input interface and the output interface to determine a first textual representation of the first audio input and transmit a first message stream comprising the first textual representation and a first identifier associated with the first audio input via the output interface, and a second recognition module coupled to the second input interface and the output interface to determine a second textual representation of the second audio input and transmit a second message stream comprising the second textual representation and a different identifier associated with the second audio input via the output interface.
1. A system comprising:
a first splitter coupled to a first audio input device, an intercom system and a first input interface to provide a first audio signal received from the first audio input device to each of the intercom system and the first input interface;
a second splitter coupled to a second audio input device, the intercom system and a second input interface to provide a second audio signal received from the second audio input device to each of the intercom system and the second input interface, wherein the intercom system is configured to combine the first audio signal and the second audio signal into a consolidated audio signal;
a first analog-to-digital converter (ADC) coupled to the first input interface to convert the first audio signal into a first input corresponding to the first audio signal;
a second analog-to-digital converter (ADC) coupled to the second input interface to convert the second audio signal into a second input corresponding to the second audio signal;
a shared output interface to be coupled to a communications bus;
a first recognition module coupled to the first ADC and the shared output interface to determine a first textual representation of the first audio signal based on the first input and transmit a first message stream comprising the first textual representation and a first identifier associated with the first audio signal via the shared output interface; and
a second recognition module coupled to the second ADC and the shared output interface to determine a second textual representation of the second audio signal based on the second input and transmit a second message stream comprising the second textual representation and a second identifier associated with the second audio signal via the shared output interface, wherein the second identifier is different from the first identifier.
2. The system of claim 1 , wherein:
the first recognition module broadcasts the first message stream comprising the first textual representation and the first identifier associated with the first audio signal over the communications bus; and
the second recognition module broadcasts the second message stream comprising the second textual representation and the second identifier associated with the second audio signal over the communications bus.
3. The system of claim 1 , further comprising a first user input device coupled to the first recognition module, wherein the first recognition module is configurable to initiate sampling the first audio signal in response to a first user input received via the first user input device.
4. The system of claim 3 , further comprising a second user input device coupled to the second recognition module, wherein:
the second user input device is different from the first user input device; and
the second recognition module is configurable to initiate sampling the second audio signal in response to a second user input received via the second user input device.
5. The system of claim 1 , further comprising a data storage element to maintain subscription information in response to subscription requests from one or more onboard systems coupled to the shared output interface via the communications bus.
6. The system of claim 5 , wherein:
the first recognition module broadcasts the first message stream comprising the first textual representation and the first identifier associated with the first audio signal over the communications bus; and
a first onboard application at the one or more onboard systems subscribed to the first identifier is configurable to receive the first message stream and initiate one or more actions associated with a voice command corresponding to the first textual representation.
7. The system of claim 6 , wherein:
the second recognition module broadcasts the second message stream comprising the second textual representation and the second identifier associated with the second audio signal over the communications bus; and
a second onboard application at the one or more onboard systems subscribed to the second identifier is configurable to receive the second message stream and initiate one or more actions associated with a second voice command corresponding to the second textual representation.
8. The system of claim 1 , wherein the first audio signal and the second audio signal are concurrently received, and the first recognition module determines the first textual representation of the first audio signal independent of the second recognition module determining the second textual representation of the second audio signal.
9. The system of claim 1 , wherein the first audio input device is associated with a first user and the second audio input device is associated with a second user different from the first user.
10. The system of claim 1 , wherein the consolidated audio signal is reproduced via one or more audio output devices.
11. The system of claim 3 , wherein the consolidated audio signal is reproduced via an audio output device integrated with the first user input device at a headset.
12. The system of claim 1 , wherein the first identifier uniquely identifies the first audio input device and the second identifier uniquely identifies the second audio input device.
13. The system of claim 1 , further comprising an avionics system coupled to the shared output interface via the communications bus, wherein a destination software application at the avionics system is configured to disambiguate a voice command corresponding to the first textual representation concurrent to a second voice command corresponding to the second textual representation using the first identifier.
14. The system of claim 1 , further comprising the intercom system including a first audio input interface coupled to the first splitter to receive the first audio signal and a second audio input interface coupled to the second splitter to receive the second audio signal.
15. A non-transitory computer-readable medium having computer-executable instructions stored thereon that, when executed by a processing system, cause the processing system to provide a speech recognition system comprising:
a first recognition module coupled to a first analog-to-digital converter (ADC) coupled to a first splitter to obtain first audio input corresponding to a first voice command comprising a first audio signal received via a first audio input device coupled to the first splitter, determine a first textual representation of the first voice command based on the first audio input, and broadcast, via a shared output interface coupled to a communications bus, a first message stream comprising the first textual representation of the first voice command in association with a first identifier associated with the first audio input device; and
a second recognition module coupled to a second analog-to-digital converter (ADC) coupled to a second splitter to obtain second audio input corresponding to a second voice command comprising a second audio signal received via a second audio input device coupled to the second splitter, wherein the second audio input device is different from the first audio input device, determine a second textual representation of the second voice command based on the second audio input, and broadcast, via the shared output interface coupled to the communications bus, a second message stream comprising the second textual representation of the second voice command in association with a second identifier associated with the second audio input device, wherein the second identifier is different from the first identifier.
16. The computer-readable medium of claim 15 , wherein the computer-executable instructions cause the processing system to determine the first identifier based at least in part on the first audio input device in response to a first subscription request from an avionics system coupled to the communications bus.
17. The computer-readable medium of claim 15 , wherein the first audio input device is associated with a first user and the second audio input device is associated with a second user different from the first user.
18. The computer-readable medium of claim 15 , wherein the first recognition module determines the first textual representation concurrent to the second recognition module determining the second textual representation.
19. A method comprising:
receiving, via a first input interface of a speech recognition system corresponding to a first speech recognition channel distinct from a second speech recognition channel associated with a second input interface of the speech recognition system, a first audio signal from a first audio input device coupled to the first input interface, wherein a second audio input device different from the first audio input device is coupled to the second input interface;
determining, at the speech recognition system, a first textual representation of the first audio signal;
determining, at the speech recognition system, a first identifier associated with the first audio input device, wherein the first identifier uniquely identifies the first audio input device and is different from a second identifier associated with the second audio input device coupled to the second input interface of the speech recognition system; and
transmitting, by the speech recognition system via a shared output interface of the speech recognition system coupled to a communications bus, a first message stream including the first identifier uniquely identifying the first audio input device, wherein the first message stream comprises the first textual representation.
20. The method of claim 19 , further comprising:
receiving, at the speech recognition system, a subscription request for the first audio input device from a system onboard a vehicle, wherein the system is coupled to the shared output interface of the speech recognition system; and
providing, by the speech recognition system, the first identifier to the system in response to the subscription request, wherein the system is configurable to monitor the shared output interface to receive the first message stream using the first identifier.