Narrator
Read it, record it, score it. All on one Mac.
Our in-house audio studio. Drop in any document and it reads it aloud, or write a script and it records a multi-voice audiobook or podcast. It also writes songs, coaches karaoke, plays MIDI through an orchestra and makes faces talk, all running locally at no cost per use.
Music Studio: pick a take, learn it, record it, tune and save.
Why It Exists
Voice and music services charge by the minute and keep your files. Narrator runs entirely on our own hardware, so the studio voiced Santa Cheer’s songs and our narrations for nothing.
- 745
- Voices
// Built with
- Runs on-device
- Kokoro TTS
- macOS Vision OCR
- Python + FastAPI
- $0 per use
The Best Parts
What Narrator does well.
01
Reads anything aloud
PDFs, scans with OCR, EPUB, Kindle, Office files and web pages, cleaned up and read sentence by sentence.
02
Audiobook and podcast studios
Multi-voice scripts with a cast, AI co-hosts, music beds and chapter breaks, exported as finished audio.
03
745 voices, plus your own
A library of voices and new ones made from a recording or a written description.
04
Music maker and karaoke
Describe a song and get lyrics, vocals and instruments, then sing over it with a guide singer and auto-tune.
05
Orchestra and lip sync
Play any MIDI file through an orchestra, and make a face in a photo or video say anything.
Inside the Product
Real screens, captured live.
Podcast Studio: cast each voice, write or generate the script, record.
Voices: hear any line in any of 745 voices, or make a new one.
Orchestra: drop in MIDI, choose who plays each part.
More From the Lab
Built alongside Narrator.
Want a Narrator of your own?
Planned, engineered and launched by one accountable team.