Methods, systems and apparatuses for improved speech recognition and transcription
AssigneeComcast Cable Communications, LLC
InventorsRui Min, Stefan Deichmann, Hongcheng Wang, Geifei Yang
Life of this patent
Estimating the term…
Abstract
Methods, systems, and apparatuses for improved speech recognition and transcription of user utterances are described herein. A user utterance may be processed by a speech recognition computing device. One or more acoustic features associated with the user utterance may be used to determine whether one or more actions are to be performed based on a transcription of the user utterance.
Claims20
A method comprising:
- receiving, at a computing device, user utterance data associated with one or more user devices;
- determining, based on the user utterance data, one or more user utterances associated with one or more erroneous automatic speech recognition (ASR) engine transcriptions;
- and
- training, based on the one or more user utterances, an acoustic model, wherein the acoustic model is trained to generate one or more updated transcriptions to override the one or more erroneous ASR engine transcriptions.
An apparatus comprising:
- one or more processors;
- and
- memory storing processor-executable instructions that, when executed by the one or more processors, cause the one or more processors to:
- receive user utterance data associated with one or more user devices;
- determine, based on the user utterance data, one or more user utterances associated with one or more erroneous automatic speech recognition (ASR) engine transcriptions;
- and
- train, based on the one or more user utterances, an acoustic model, wherein the acoustic model is trained to generate one or more updated transcriptions to override the one or more erroneous ASR engine transcriptions.
One or more non-transitory computer-readable media storing processor-executable instructions thereon, which, when executed by at least one processor, cause the at least one processor to:
- receive user utterance data associated with one or more user devices;
- determine, based on the user utterance data, one or more user utterances associated with one or more erroneous automatic speech recognition (ASR) engine transcriptions;
- and
- train, based on the one or more user utterances, an acoustic model, wherein the acoustic model is trained to generate one or more updated transcriptions to override the one or more erroneous ASR engine transcriptions.
Dependencies are inferred from the claim text.