Virtual assistant configured by selection of wake-up phrase

• US 10,217,453 B2
• Filed: 10/14/2016
• Issued: 02/26/2019
• Est. Priority Date: 10/14/2016
• Status: Active Grant

• Current Assignee: Soundhound AI IP LLC (SoundHound AI, Inc. (f/k/a Archimedes Tech SPAC Partners Co.))
• Original Assignee: SoundHound, Inc. (SoundHound AI, Inc. (f/k/a Archimedes Tech SPAC Partners Co.))
• Inventors: Stevans, Mark, Almudafar-Depeyrot, Monika, Mohajer, Keyvan
• Primary Examiner(s): Yen, Eric
• Application Number:
  US15/294,234
• Publication Number:
  US 20180108343A1
• Time in Patent Office: 865 Days
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 2015/025 Phonemes, fenemes or fenones being the recognition units
• CPC Class Codes: G10L 2015/088 Word spotting
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command

Abstract

A speech-enabled dialog system responds to a plurality of wake-up phrases. Based on which wake-up phrase is detected, the system'"'"'s configuration is modified accordingly. Various configurable aspects of the system include selection and morphing of a text-to-speech voice; configuration of acoustic model, language model, vocabulary, and grammar; configuration of a graphic animation; configuration of virtual assistant personality parameters; invocation of a particular user profile; invocation of an authentication function; and configuration of an open sound. Configuration depends on a target market segment. Configuration also depends on the state of the dialog system, such as whether a previous utterance was an information query.

Claims

1. A method of configuring a computerized dialog system, the method comprising:

receiving a request including an indication of which of a plurality of wake-up phrases has been detected;

identifying a knowledge domain associated with the detected wake-up phrase from the plurality of wake-up phrases; and

configuring a text-to-speech (TTS) system, responsive to receiving the request, to use the identified knowledge domain to respond to the request.

2. The method of claim 1 wherein the step of configuring includes assigning speech morphing parameters.

3. The method of claim 1 further comprising configuring a language model.

4. The method of claim 1 further comprising configuring an ASR acoustic model.

5. The method of claim 1 further comprising configuring a natural language grammar.

6. The method of claim 1 further comprising configuring a graphic animation.

7. The method of claim 1 further comprising configuring personality parameters.

8. The method of claim 1 further comprising invoking a particular user profile.

9. The method of claim 1 further comprising participating in an authentication function.

10. The method of claim 1 further comprising configuring an open sound.

11. The method of claim 1 wherein the step of configuring varies depending on the state of the dialog system.

12. The method of claim 11 wherein a TTS voice is unavailable in a particular state.

13. A non-transitory computer readable medium storing code that, if executed by one or more computers would cause the one or more computers to:

receive an indication of which of a plurality of wake-up phrases associated with a virtual agent has been detected;

identify a knowledge domain associated with the detected wake-up phrase from the plurality of wake-up phrases; and

configure a text-to-speech (TTS) system to use the identified knowledge domain to respond to a request.

14. A system for hosting virtual assistant plugins, the system comprising:

a digital storage medium for storing a plurality of a text-to-speech (TTS) voices;

a network interface enabled to receive indications of which of a plurality of wake-up phrases has been detected; and

a processing device enabled to identify a knowledge domain associated with the detected wake-up phrase and configure one of the plurality of TTS voices based on a received indication to use the identified knowledge domain to respond to a request.

15. A voice-enabled device comprising at least one non-transitory computer readable medium storing code that, when executed by one or more processors, would cause the device to:

spot for a plurality of wake-up phrases; and

responsive to detecting a wake-up phrase from the plurality of wake-up phrases;

retrieve open sound audio data;

identify a knowledge domain associated with the detected wake-up phrase;

configure the device to use the identified knowledge domain to respond to requests; and

output the open sound audio data.

16. The device of claim 15, wherein the open sound audio data is stored by the device.

17. The device of claim 15 wherein the retrieving comprises sending to a server an indication of which of the wake-up phrases was detected and the open sound audio data is retrieved from the server.

18. A non-transitory computer readable medium storing code that, if executed by one or more computers, would cause the one or more computers to:

associate one wake-up phrase with each of a plurality of plugins resulting in a plurality of wake-up phrases;

spot one of the plurality of wake-up phrases;

invoke a first plugin associated with the spotted wake-up phrase, wherein the first invoked plugin acts as a virtual assistant

identify a knowledge domain associated with the spotted wake-up phrase;

configure the first invoke plugin to use the identified knowledge domain to cause the one our more computers to respond to requests.

19. The non-transitory computer readable medium of claim 18, wherein invoking the first plugin causes the one or more computers to execute code that accesses a web API that provides the functionality of the virtual assistant.

20. The non-transitory computer readable medium of claim 18, wherein the one or more computers is caused to:

spot a second one of the plurality of wake-up phrases; and

invoke a second plugin associated with second spotted wake-up phrase, wherein the second invoked plugin is different from the first invoked plugin.