transcribe

To transcribe audio or video into text, provide a valid absolute file path to the media file. The tool converts video to WAV audio, chunks large files, and processes speech-to-text using your configured provider.

You can transcribe video files directly without manual conversion. The tool automatically extracts audio from video, normalizes it to mono 16kHz WAV, and splits large media files into chunks for processing.

Media transcription supports multiple speech-to-text providers including OpenAI Whisper, Deepgram, and Google Gemini. The active provider is selected via your configured settings.

Transcription requires a valid absolute file path to locate and process the media. Relative paths or unsupported file extensions are rejected by the system to prevent processing errors.

If a speech-to-text provider is not configured, the transcription process cannot run and returns a clear error. You must configure an STT provider in settings before attempting media transcription.

For large media files like long podcasts, the tool normalizes the audio to WAV format and automatically splits the file into smaller chunks. Each chunk is processed sequentially by the speech-to-text provider.