Back to results
US 10,832,654 B2Granted
Recognizing accented speech
AssigneeGoogle Technology Holdings LLC
InventorKristin A. Gray
Life of this patent
Estimating the term…
Abstract
Techniques (300, 400, 500) and apparatuses (100, 200, 700) for recognizing accented speech are described. In some embodiments, an accent module recognizes accented speech using an accent library based on device data, uses different speech recognition correction levels based on an application field into which recognized words are set to be provided, or updates an accent library based on corrections made to incorrectly recognized speech.
Claims20
A computer-implemented method comprising:
- receiving, by an automated speech recognition system that includes a speech recognition engine, audio data of an utterance;
- selecting, by the automated speech recognition system, a linguistic library that includes the words of a language;
- identifying, by the automated speech recognition system, a native language of a speaker of the utterance based on tones or intonation within the audio data of the utterance;
- based on the identified native language of the speaker of the utterance, selecting, by the automated speech recognition system, two or more accent libraries that each include phonemes for different pronunciations for the words of the language;
- generating, by the automated speech recognition system, a combined accent library by combining the two or more accent libraries;
- obtaining a transcription of the utterance by performing, by the speech recognition engine, speech recognition on the audio data of the utterance using the linguistic library and the combined accent library;
- and
- providing, for output, the transcription of the utterance.
5 more elements
A system comprising:
- one or more computers;
- and
- one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
- receiving, by an automated speech recognition system that includes a speech recognition engine, audio data of an utterance;
- selecting, by the automated speech recognition system, a linguistic library that includes the words of a language;
- identifying, by the automated speech recognition system, a native language of a speaker of the utterance based on tones or intonation within the audio data of the utterance;
- based on the identified native language of the speaker of the utterance, selecting, by the automated speech recognition system, two or more accent libraries that each include phonemes for different pronunciations for the words of the language;
- generating, by the automated speech recognition system, a combined accent library by combining the two or more accent libraries;
- obtaining a transcription of the utterance by performing, by the speech recognition engine, speech recognition on the audio data of the utterance using the linguistic library and the combined accent library;
- and
- providing, for output, the transcription of the utterance.
8 more elements
A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:
- receiving, by an automated speech recognition system that includes a speech recognition engine, audio data of an utterance;
- selecting, by the automated speech recognition system, a linguistic library that includes the words of a language;
- identifying, by the automated speech recognition system, a native language of a speaker of the utterance based on tones or intonation within the audio data of the utterance;
- based on the identified native language of the speaker of the utterance, selecting, by the automated speech recognition system, two or more accent libraries that each include phonemes for different pronunciations for the words of the language;
- generating, by the automated speech recognition system, a combined accent library by combining the two or more accent libraries;
- obtaining a transcription of the utterance by performing, by the speech recognition engine, speech recognition on the audio data of the utterance using the linguistic library and the combined accent library;
- and
- providing, for output, the transcription of the utterance.
5 more elements
Dependencies are inferred from the claim text.