transcribe

To transcribe a meeting recording, provide the absolute file path of your mp4 video. The system validates the extension, optionally uses ffmpeg to extract and normalize audio to 16kHz mono WAV, and routes it through your configured speech-to-text provider.

Yes, large audio files are handled safely by automatically normalizing media to 16kHz mono WAV, splitting large inputs into smaller chunks, transcribing each chunk via your configured STT provider like OpenAI Whisper, and combining the results.

Speech-to-text transcription supports common audio formats like mp3, wav, and m4a, alongside video formats such as mp4 and mov. The system validates file extensions before processing the media into text output.

FFmpeg is used for optional audio extraction and conversion to 16kHz mono WAV. It processes media files for transcription, but the core requirement is configuring a speech-to-text provider in your settings services.

Speech-to-text transcription of long recordings uses automatic chunking to split large inputs. It transcribes each chunk individually using the configured STT provider and then combines the text results.

The best way to convert recorded video into text is providing the absolute file path. The system validates the media, extracts audio using ffmpeg, and routes it through your selected speech-to-text provider to return a readable transcript.