Speech recognition using phoneme matching
A system, method and computer program is provided for generating customized text representations of audio commands. A first speech recognition module may be used for generating a first text representation of an audio command based on a general language grammar. A second speech recognition module may be used for generating a second text representation of the audio command, the second module including a custom language grammar that may include contacts for a particular user. Entity extraction is applied to the second text representation and the entities are checked against a file containing personal language. If the entities are found in the user-specific language, the two text representations may be fused into a combined text representation and named entity recognition may be performed again to extract further entities.
1. A computer implemented method of converting an audio input into a text representation associated with a user, the method comprising:
receiving, by a conversion processor, the audio input;
generating, by a first speech recognition processor, a first text representation of the audio input, wherein the first speech recognition processor uses a first natural language model for speech recognition of a first language;
generating, by a second speech recognition processor, a second text representation of the audio input using a second natural language model, wherein the second natural language model recognizes a word not in the first natural language model for speech recognition, the second text representation includes the word not in the first natural language model, the second text representation includes a phoneme of the word, and the second natural language model is distinct from the first natural language model;
aligning, by the conversion processor, based on a phoneme sequence associated with the first and second text representations, the first text representation and the second text representation;
generating, by the conversion processor, based at least on the aligned first and second text representations and a likelihood of the phoneme of the word being a part of the first text representation, a third text representation; and
outputting the third text representation as a personalized recognized text representation of the audio input.
2. The method recited in claim 1 , wherein the word comprises a location name associated with a location, the location being within a predetermined distance from an audio input device associated with the audio input.
3. The method recited in claim 1 , wherein the audio input is created in response to an audio signal received at a microphone.
4. The method recited in claim 1 , wherein the word comprises a contact name from an address book.
5. The method recited in claim 1 , further comprising substituting a portion of the first text representation with the word to create the second text representation of the audio input.
6. The method recited in claim 1 , wherein the audio input is received at an automatic speech recognition system from an application, wherein the application provides an interface for receiving the audio input.
7. A computer system for converting an audio input into a recognized text representation, the computer system comprising processors configured to execute operations comprising:
receiving, by a conversion processor, the audio input;
generating, by a first speech recognition processor, a first text representation of the audio input using a first natural language model for speech recognition of a first language;
substituting a portion of the first text representation with a word to create a second text representation of the audio input, wherein the word is associated with a second language, wherein the second language is distinct from the first language;
aligning phonetics of the first text representation and the second text representation;
generating, based at least on the aligned phonetics of the first and second text representation and a likelihood of a phoneme of the word being a part of the first text representation, a third text representation; and
outputting the third text representation as a personalized recognized text representation of the audio input.
8. The computer system of claim 7 , further comprising a processor configured to execute operations comprising:
returning an indication whether the word is present in the first text representation, wherein the indication indicates a location of the word within the first text representation.
9. The computer system of claim 7 , wherein the aligning further comprises providing an indication whether the word is present in the first text representation, wherein the indication indicates a location of the word within the first text representation.
10. The computer system of claim 7 , further comprising a processor configured to execute operations comprising:
creating a first phonetic representation of the word and a second phonetic representation of the first text representation; and
comparing compare the first phonetic representation and the second phonetic representation to determine a likelihood that the word is present in the first text representation.
11. The computer system of claim 7 , the system further comprising:
an input device configured to generate the audio input for an automatic speech recognition system in response to input received at the input device, wherein the input device is a microphone for receiving an audio signal, and wherein the audio input is created in response to the audio signal.
12. The computer system of claim 7 , wherein the word comprises a contact name from an address book.
13. The computer system of claim 7 , wherein the word comprises a location name associated with a location, the location being within a predetermined distance from an input device.
14. The computer system of claim 7 , wherein the word is selected from a plurality of common words.
15. The computer system of claim 7 , the system further comprising:
an application configured to provide an interface for receiving the audio input.
16. The computer system of claim 10 , the system further comprising: transferring the word to create the first phonetic representation of the word and the second phonetic representation of the first text representation.
17. A device for converting an audio input into a recognized text representation associated with a user, the device comprising a processor configured to execute operations comprising:
receiving, by a conversion processor, the audio input;
generating, by a first speech recognition processor, a first text representation of the audio input, wherein the first speech recognition processor uses a first natural language model for speech recognition of a first language;
generating, by a second speech recognition processor, a second text representation of the audio input using a second natural language model, wherein the second natural language model recognizes a word not in the first natural language model for speech recognition, the second text representation includes the word not in the first natural language model, the second text representation includes a phoneme of the word, and the second natural language model is distinct from the first natural language model;
aligning, by the conversion processor based on a phoneme sequence associated with the first and second text representation, the first text representation and the second text representation;
generating, based at least on the aligned first and second text representations and a likelihood of the phoneme of the word being a part of the first text representation, a third text representation; and
outputting the third text representation as a personalized recognized text representation of the audio input.
18. The device of claim 17 , wherein the audio input is created in response to an audio signal received at a microphone.
19. The device of claim 17 , wherein the word comprises a contact name from an address book.
20. The device of claim 17 , the processor further configured to execute operations comprising:
substituting a portion of the first text representation with the word to create the second text representation of the audio input.