Turn audio & video into clean, speaker-labeled transcripts

Upload a file, paste a YouTube link, or drop in text. Get transcripts, speaker separation, summaries, translations and downloadable reports.

Drop an audio or video file, or click to browse

MP4, MOV, MKV, WEBM · WAV, MP3, M4A, FLAC, OGG

file

YouTube URL

Simple mode prefers existing captions; speaker mode always downloads and analyzes the audio.

Drop a .txt file, or click to browse

Meeting notes, articles, transcripts — we'll summarize it

file

Mode

Fastest path: transcript & summary only.

Translation (optional)

Adds a translated summary and per-line translations.

Processing time: on a CPU-only server expect roughly 1–3 minutes per minute of media. Speaker separation is slower. Keep clips short while testing.

Starting…

Your file is being processed. This can take a few minutes.

Accurate transcription

Local Whisper with word-level timestamps.

Speaker separation

pyannote 3.1 with a local CPU fallback.

Translation

Summaries & lines in 10+ languages.

Exports

.txt, .srt and polished PDF reports.