Transcription & voice studio
Every word in its place: speakers, timestamps, subtitles, summaries and voiceovers for podcasts, interviews, courses and support calls.
Thanks for having me. I used to rush every morning, and it made the whole day feel loud.
Drop in audio or video, paste a link, or record in the browser. Minutes are reserved up front and returned if a file fails.
Every word has a start and end time, a confidence shade and a speaker. Rename a speaker once and every line follows.
Click a word to jump to it. Play and the current word lights up. Find and replace keeps timings; every save is a revision.
SRT and VTT with 42-character lines, two lines per cue and clean timing. Translate a transcript and its subtitles together.
Summaries, chapters, key points, action items and quotes, each labelled while the built-in simulated engine is active.
Find “renewal clause” across all your calls and open the exact second it was said.
Turn a script into audio with several voices, speeds and pauses. Export WAV, or MP3 when ffmpeg is installed.
Workspaces, roles, comments on words, and read-only share links with a player that expire and can be revoked.
Scoped keys, an OpenAPI 3.1 spec, signed webhooks and idempotent uploads for your own automation.
Connect a Whisper-compatible server or an OpenAI-compatible endpoint with your own key. Nothing is called until you switch it on.
Minute packs that never expire are available on every plan. Prices are what you charge your own customers; the operator edits them in the plan table.
For solo podcasters and researchers
$19/mo
For teams
$59/mo
For agencies and support desks
$199/mo
Out of the box the product runs a built-in simulated engine that returns sample results, clearly labelled. To get real transcripts, summaries and voices, connect your own provider account and key. No AI usage, credits or keys are included; provider costs are yours.
MP3, M4A, WAV, OGG, WebM and MP4 work as they are. MOV, MKV, FLAC and WMA need ffmpeg installed on your server.
One minute per minute of audio, rounded up for each file. Voiceovers cost one minute per 1,000 characters. Plan minutes are spent first; minute packs never expire.
Yes. Every screen works with the simulated engine, including the editor, subtitles, translation, search and voiceovers.