System and method for combining speech recognition outputs from a plurality of domain-specific speech recognizers via machine learning
View Patent ↗Disclosed herein are systems, methods and non-transitory computer-readable media for performing speech recognition across different applications or environments without model customization or prior knowledge of the domain of the received speech. The disclosure includes recognizing received speech with a collection of domain-specific speech recognizers, determining a speech recognition confidence for each of the speech recognition outputs, selecting speech recognition candidates based on a respective speech recognition confidence for each speech recognition output, and combining selected speech recognition candidates to generate text based on the combination.
1. A method comprising:
recognizing, via a processor, received speech with a plurality of domain-specific speech recognizers without knowledge of a domain of the received speech, the plurality of domain-specific speech recognizers comprising two domain-specific speech recognizers from different domains and two domain-specific speech recognizers from a specific domain, wherein each domain-specific speech recognizer of the plurality of domain-specific speech recognizers recognizes the received speech, to yield respective speech recognition outputs;
determining a speech recognition confidence score for each of the respective speech recognition outputs;
selecting speech recognition candidates from segments of the respective speech recognition outputs based on the speech recognition confidence score for the respective speech recognition outputs;
combining, via a machine-learning algorithm, the speech recognition candidates, to yield a combination of the speech recognition candidates; and
generating text based on the combination.
2. The method of claim 1 , wherein the speech recognition candidates are selected without knowledge of the domain of the received speech.
3. The method of claim 1 , wherein one of the different domains comprises one of travel, banking, and business.
4. The method of claim 1 , wherein the machine-learning algorithm comprises a mixture of domain-specific speech recognizers from different domains, wherein the mixture of domain-specific speech recognizers comprises two of the following: local business search, web search, Short Messaging Service, question/answering, video search, broadcast news, and voicemail to text.
5. The method of claim 4 , wherein selecting the speech recognition candidates further comprises comparing domain-specific speech recognizers in the mixture of domain-specific speech recognizers to select the best speech recognition candidates.
6. The method of claim 1 , wherein selecting the speech recognition candidates further comprises selecting a speech recognition candidate having a highest confidence score.
7. The method of claim 1 , wherein selecting the speech recognition candidates further comprises:
dividing the received speech into substrings; and
selecting a best speech recognition candidate for each substring.
8. The method of claim 1 , further comprising mixing the speech recognition candidates.
9. The method of claim 1 , further comprising creating a lattice of the speech recognition candidates.
10. The method of claim 9 , further comprising:
collecting statistics based on the speech recognition candidates; and
training the machine-learning algorithm based on the statistics.
11. The method of claim 10 , wherein training parameters are based on one of a lattice combination and a neural network graph that learns from an edit distance between the speech recognition candidates and a correct recognition candidate.
12. The method of claim 1 , wherein a speech recognition candidate comprises one of a lattice, confidence scores, and speech recognition metadata.
13. The method of claim 1 , further comprising:
collecting statistics based on the speech recognition candidates; and
training parameters associated with the domain-specific speech recognizers based on the statistics.
14. A system comprising:
a processor; and
a computer-readable storage medium having instructions stored which, when executed by the processor, result in the processor performing operations comprising:
recognizing, via a processor, received speech with a plurality of domain-specific speech recognizers without knowledge of a domain of the received speech, the plurality of domain-specific speech recognizers comprising two domain-specific speech recognizers from different domains and two domain-specific speech recognizers from a specific domain, wherein each domain-specific speech recognizer of the plurality of domain-specific speech recognizers recognizes the received speech, to yield respective speech recognition outputs;
determining a speech recognition confidence score for each of the respective speech recognition outputs;
selecting speech recognition candidates from segments of the respective speech recognition outputs based on the speech recognition confidence score for the respective speech recognition outputs;
combining, via a machine-learning algorithm, the speech recognition candidates, to yield a combination of the speech recognition candidates;
generating text based on the combination;
collecting usage statistics based on the speech recognition candidates;
training parameters associated with the plurality of domain-specific speech recognizers based on the usage statistics, to yield training parameters;
and
training the machine-learning algorithm based on the usage statistics and the training parameters.
15. The system of claim 14 , wherein the speech recognition candidates are selected without knowledge of the domain of the received speech.
16. The system of claim 14 , wherein the training parameters for the machine-learning algorithm are based on one of a lattice combination and a neural network graph that learns from an edit distance between the speech recognition candidates and a correct recognition candidate.
17. A computer-readable storage device having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:
recognizing, via a processor, received speech with a plurality of domain-specific speech recognizers without knowledge of a domain of the received speech, the plurality of domain-specific speech recognizers comprising two domain-specific speech recognizers from different domains and two domain-specific speech recognizers from a specific domain, wherein each domain-specific speech recognizer of the plurality of domain-specific speech recognizers recognizes the received speech, to yield respective speech recognition outputs;
determining a speech recognition confidence score for each of the respective speech recognition outputs;
selecting speech recognition candidates from segments of the respective speech recognition outputs based on the speech recognition confidence score for the respective speech recognition outputs;
combining, via a machine-learning algorithm, the speech recognition candidates, to yield a combination of the speech recognition candidates; and
generating text based on the combination.
18. The computer-readable storage device of claim 17 , wherein the speech recognition candidates are selected without knowledge of the domain of the received speech.
19. The computer-readable storage device of claim 17 , wherein training parameters for the machine-learning algorithm are based on one of a lattice combination and a neural network graph that learns from an edit distance between the speech recognition candidates and a correct recognition candidate.