Zoom Video Communications, Inc. (20250078829). AUTOMATIC SPEECH RECOGNITION USING MULTIPLE LANGUAGE MODELS

AUTOMATIC SPEECH RECOGNITION USING MULTIPLE LANGUAGE MODELS

Organization Name

Zoom Video Communications, Inc.

Inventor(s)

Thai Son Nguyen of Karlsruhe (DE)

Jie Pu of Baden-Wurttemberg (DE)

[[:Category:Sebastian St�ker of Karlsruhe (DE)|Sebastian St�ker of Karlsruhe (DE)]][[Category:Sebastian St�ker of Karlsruhe (DE)]]

AUTOMATIC SPEECH RECOGNITION USING MULTIPLE LANGUAGE MODELS

This abstract first appeared for US patent application 20250078829 titled 'AUTOMATIC SPEECH RECOGNITION USING MULTIPLE LANGUAGE MODELS

Original Abstract Submitted

one example method includes receiving an audio stream comprising speech; generating, by automatic speech recognition (“asr”) software, a plurality of hypotheses, each hypothesis comprising a transcription of a first portion of the speech; rescoring, using a first trained language model, each hypothesis of the plurality of hypotheses; and responsive to a first hypothesis not satisfying a threshold, generating and outputting, using a trained large language model (“llm”), a final transcription based on the plurality of hypotheses.