Nodes/MisoTTS-ComfyUI/Miso TTS - Whisper Transcribe
ComfyUI Node

Miso TTS - Whisper Transcribe

Transcribe ComfyUI AUDIO with Whisper and output transcript text for reference_text.

By Saganaki22·Created 3 months ago·Updated 3 months ago· 7
Miso TTS - Whisper Transcribe
  • audio
  • transcript
modelwhisper-large-v3-turbo (auto-download)
dtypeauto
languageauto
tasktranscribe
chunk_length_s30
download_if_missingtrue
CategoryMiso TTS

Inputs (7)

NameTypeDefaultDescription
audioAUDIOAudio to transcribe. Connect this output to Miso reference_text.
modelCOMBOwhisper-large-v3-turbo (auto-download)Whisper ASR model. Auto-download options are stored under ComfyUI/models/audio_encoders/.
dtypeCOMBOautoWhisper precision. auto uses bf16/fp16 on GPU and fp32 on CPU.
languageCOMBOautoOptional language hint for Whisper.
taskCOMBOtranscribetranscribe keeps the source language; translate outputs English.
chunk_length_sINT300–120Chunk length in seconds for longer audio. 0 disables chunking.
download_if_missingBOOLEANtrueDownload selected auto-download Whisper model if missing.

Outputs (1)

NameTypeDescription
transcriptSTRING