Speech and singing engine with sixteen voices that runs on your CPU
Stage
Not given
Not given
Founded
Not given
Not given
Based in
Not given
Not given
Pricing model
Freemium
Open source
Pricing
Free: 3 files a day, up to 30 minutes each, no credit card required. Plus: $6/month (billed $72 yearly), 1,200 minutes per month. Unlimited: $10/month (billed $120 yearly), unlimited transcription with fair use cap of 12,000 minutes per month. Minute Pack: $10 for 600 minutes or $1 per 60 minutes, s
Not given
Key features
Speaker identification with automatic labeling
Synced playback with timestamps
Support for 90+ languages
Multiple export formats including Word, PDF, TXT and SRT
Automatic language detection
Live transcription progress tracking
Audio repair for noisy recordings
Support for files up to 5 GB on paid plans
AI prompt export and translation on all plans
Secure file storage never used for training
Sixteen voices for speech and singing
Piano roll music editor with syllable-based lyrics
MIDI file import and export with lyrics
Vibrato control and voice mixing
Runs in browser and natively on macOS and Linux
Python and command line interfaces
0.1 second latency to first sound
Parts and scores saved as plain text files
What makes it different
Accurate Whisper-class speech recognition that finishes most files in minutes
Transcripts read like a script with automatic speaker labels and synced playback
Free trial with no credit card required
Handles large files up to 5 GB with automatic resumption on interrupted uploads
Simple pricing with honest limits and no surprises
Runs entirely on your CPU with no cloud dependency
12 times faster than real time on a single CPU core