Batch Transcription
A speech recognition mode in which an entire audio segment is fed to the model at once to complete transcription, prioritizing throughput, with a relatively fixed computation graph.
Timeline (last 90 days)
Streaming speech recognition is fundamentally different from batch transcription: streaming recognition requires text to appear in real time as speech is spoken, rather than waiting for the entire utterance to finish and then outputting it all at once.
The computation graph for batch transcription is relatively fixed, and optimization opportunities are concentrated in operator fusion and matrix operation acceleration.
All Facts (2)
Streaming speech recognition is fundamentally different from batch transcription: streaming recognition requires text to appear in real time as speech is spoken, rather than waiting for the entire utterance to finish and then outputting it all at once.
50%UnverifiedThe computation graph for batch transcription is relatively fixed, and optimization opportunities are concentrated in operator fusion and matrix operation acceleration.
50%