daanelson
/
whisperx
Accelerated transcription of audio using WhisperX
Run daanelson/whisperx with an API
Input schema
audiouri
Audio file
debugboolean
Print out memory usage information.
only_textboolean
Set if you only want to return text; otherwise, segment metadata will be returned as well.
batch_sizeinteger
Parallelization of input audio transcription
- Default
- 32
align_outputboolean
Use if you need word-level timing and not just batched transcription. Only works for English atm
Output schema
{
"title": "Output"
}