Fish Audio: Speech to Text
Transcribe audio to text with Fish Audio ASR. Language is detected automatically across 80+ languages, with optional word-level timestamps returned as JSON segments. Use it to create subtitles, transcribe recordings, or generate text prompts from audio clips. A sample audio is pre-loaded so you can test it right away.
About
This ComfyUI tutorial workflow demonstrates end-to-end speech-to-text using the FishAudioSpeechToText node. Audio is loaded with LoadAudio (a sample file, fish_audio_example.mp3, is pre-connected so you can run it immediately), then sent to the Fish Audio ASR service for transcription. The node automatically detects the spoken language across 80+ languages and returns a clean transcript string along with an optional JSON payload of timestamped segments. The transcript is saved to disk via SaveText, while a MarkdownNote provides on-canvas tips.
What you provide
- Reading the workflow
Runs in ComfyUI. This workflow can't run on this page. Open it in Comfy Cloud with everything loaded, or download it for your own ComfyUI.
Try in CloudQuestions & answers
Enable precise_timestamps in the FishAudioSpeechToText node. This fills segments_json with per-segment text and start/end times (and, when available, finer word-level timing).
The node auto-detects language across 80+ languages. You don’t need to set the language manually—just provide clear audio.
SaveText writes the main transcript string to your ComfyUI output folder. Check the SaveText node’s path setting to see or change the exact location.
Yes. FishAudioSpeechToText calls an external ASR service, so a network connection is required. If your installation requires credentials, configure them as instructed for your environment.