Low resource key phrase detection for wake on voice

Patent metadata

Field	Value
Publication number	US-9792907-B2
Application number	US-201514950670-A
Country	US
Kind code	B2
Filing date	Nov 24, 2015
Priority date	Nov 24, 2015
Publication date	Oct 17, 2017
Grant date	Oct 17, 2017

Abstract

Techniques related to key phrase detection for applications such as wake on voice are discussed. Such techniques may include updating a start state based rejection model and a key phrase model based on scores of sub-phonetic units from an acoustic model to generate a rejection likelihood score and a key phrase likelihood score and determining whether received audio input is associated with a predetermined key phrase based on the rejection likelihood score and the key phrase likelihood score.

First claim

What is claimed is:

1. A computer-implemented method for key phrase detection comprising:
generating, via acoustic scoring of an acoustic model, a time series of scores of sub-phonetic units based on a time series of feature vectors representative of received audio input;
updating a start state based rejection model and a key phrase model associated with a predetermined key phrase based on at least some of the time series of scores of sub-phonetic units to generate a rejection likelihood score and a key phrase likelihood score, wherein the start state based rejection model consists of a single rejection state and comprises a plurality of rejection model self loops each associated with a particular score of the scores of sub-phonetic units of the acoustic model, wherein the key phrase model comprises a plurality of key phrase states interconnected by transitions therebetween, the plurality of key phrase states each comprising a self loop associated with a particular score of the scores of sub-phonetic units of the acoustic model, and wherein the start state based rejection model and the key phrase model are connected by a first transition from the single rejection state to a first key phrase state of the plurality of key phrase states; and
determining whether the received audio input is associated with the predetermined key phrase based on the rejection likelihood score and the key phrase likelihood score.

2. The method of claim 1 , wherein updating the start state based rejection model and the key phrase model comprises:
providing a continual summing at the single rejection state of the start state based rejection model based on a previous score of the single rejection state and the particular scores corresponding to the plurality of rejection model self loops; and
providing a continual summing at each key phrase state of the plurality of key phrase states based on a previous score of each key phrase state, the particular score corresponding to the self loop of each key phrase state, and a second score transitioned to each key phrase state from another state.

3. The method of claim 2 , wherein updating the start state based rejection model and the key phrase model further comprises:
comparing a sum of the previous score for a first key phrase state and a particular score corresponding to the self loop of the first key phrase state to a score for a second key phrase state interconnected to the first key phrase state by a first transition; and
updating the score for the second key phrase state to the sum when the sum is greater than the score for the second key phrase state.

4. The method of claim 1 , wherein the key phrase model comprises a multi-state lexicon look up key phrase model and the transitions of the key phrase model are associated with the lexicon look up for the predetermined key phrase.

5. The method of claim 4 , wherein the key phrase likelihood score is associated with a final state of the multi-state lexicon look up key phrase model.

6. The method of claim 1 , wherein determining whether the received audio input is associated with the predetermined key phrase comprises:
determining a log likelihood score based on the rejection likelihood score and the key phrase likelihood score; and
comparing the log likelihood score to a threshold.

7. The method of claim 1 , wherein the acoustic model comprises a deep neural network and the time series of feature vectors comprises a first feature vector comprising a stack of a time series of coefficients each associated with a sampling time.

8. The method of claim 1 , further comprising:
updating a second key phrase model associated with a second predetermined key phrase based on at least some of the time series of scores of sub-phonetic units to generate a second key phrase likelihood score; and
determining whether the received audio input is associated with the second predetermined key phrase based on the rejection likelihood score and the second key phrase likelihood score.

9. The method of claim 8 , wherein the received audio input is associated with the second predetermined key phrase, the method further comprising:
providing a system command corresponding to the second predetermined key phrase.

10. A system for performing key phrase detection comprising:
a memory configured to store an acoustic model, a start state based rejection model, and a key phrase model associated with a predetermined key phrase; and
a digital signal processor coupled to the memory, the digital signal processor to generate, based on the acoustic model, a time series of scores of sub-phonetic units based on a time series of feature vectors representative of an audio input, to update the start state based rejection model and the key phrase model based on at least some of the time series of scores of sub-phonetic units to generate a rejection likelihood score and a key phrase likelihood score, wherein the start state based rejection model consists of a single rejection state and comprises a plurality of rejection model self loops each associated with a particular score of the scores of sub-phonetic units of the acoustic model, wherein the key phrase model comprises a plurality of key phrase states interconnected by transitions therebetween, the plurality of key phrase states each comprising a self loop associated with a particular score of the scores of sub-phonetic units of the acoustic model, and wherein the start state based rejection model and the key phrase model are connected by a first transition from the single rejection state to a first key phrase state of the plurality of key phrase states, and to determine whether the received audio input is associated with the predetermined key phrase based on the rejection likelihood score and the key phrase likelihood score.

11. The system of claim 10 , wherein the digital signal processor to update the start state based rejection model and the key phrase model comprises the digital signal processor to provide a continual summing at the single rejection state of the start state based rejection model based on a previous score of the single rejection state and the particular scores corresponding to the plurality of rejection model self loops and to provide a continual summing at each key phrase state of the plurality of key phrase states based on a previous score of each key phrase state, the particular score corresponding to the self loop of each key phrase state, and a second score transitioned to each key phrase state from another state.

12. The system of claim 10 , wherein the digital signal processor to update the start state based rejection model and the key phrase model comprises the digital signal processor further to compare a sum of the previous score for a first key phrase state and a particular score corresponding to the self loop of the first key phrase state to a score for a second key phrase state interconnected to the first key phrase state by a first transition and to update the score for the second key phrase state to the sum when the sum is greater than the score for the second key phrase state.

13. The system of claim 10 , wherein the key phrase model comprises a multi-state lexicon look up key phrase model and the transitions of the key phrase model are associated with the lexicon look up for the predetermined key phrase.

14. The system of claim 13 , wherein the key phrase likelihood score is associated with a final state of the multi-state lexicon look up key phrase model.

15. The system of claim 10 , wherein the digital signal processor to determine whether the received audio input is associated with the predetermined

Assignees

• Intel Ip Corp

Inventors

• Bocklet Tobias
• Hofer Joachim

Classifications

• G06F40/284

  Lexical analysis, e.g. tokenisation or collocates · CPC title

• G10L15/142

  Hidden Markov Models [HMMs] · CPC title

• G10L15/22 Primary

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title

• G10L2015/025

  Phonemes, fenemes or fenones being the recognition units · CPC title

• G10L15/14

  using statistical models, e.g. Hidden Markov Models [HMMs] (G10L15/18 takes precedence) · CPC title