Generate speech with customizable voices in any language, and create captivating stories using natural-sounding voices.
txt2mp3
$29Starting price
0Popularity
About txt2mp3
txt2mp3 is a macOS application that generates high-quality speech from text on Apple silicon hardware. It supports voice cloning from short recordings and voice design via descriptive sentences, producing MP3 output at 48 kHz. The tool operates entirely offline, with no server uploads, accounts, or telemetry. It includes fourteen preset voices across seven accents and supports real-time generation with word-level timestamps. A local MCP server allows AI agents like Claude or Cursor to interface with the app for programmatic speech generation. The application is distributed as a one-time purchase with no recurring fees or usage limits.
Key features
- Voice cloning from short recordings
- Voice design via descriptive sentences
- 14 preset voices across 7 accents
- MP3 export at 48 kHz with word timestamps
- Local MCP server for AI agent integration
- Real-time generation with playback preview
- Offline operation after initial setup
- One-time purchase covering all v1 updates
Use cases
- Generating audiobooks or narrations locally
- Creating voiceovers for videos without cloud services
- Prototyping AI agent interactions with synthetic speech
Pros
- Runs entirely on Apple silicon Macs
- No subscription or usage metering
- Local voice cloning and design without cloud uploads
- Includes fourteen preset voices across seven accents
- Local MCP server for AI agent integration
Cons
- Requires Apple silicon Mac
- Initial download requires approximately 5.5 GB of disk space
- MCP server disabled by default and restricted to localhost
Frequently asked questions about txt2mp3
Is txt2mp3 a subscription?
No, txt2mp3 operates on a one-time payment model. A single license covers unlimited generation and all updates within version 1, with no recurring fees.
What is limited in the free trial?
The free trial restricts the amount of text the app will read at a time. All voices and features work exactly as in the paid version, and the trial does not expire.
How many Macs does one license cover?
One license can be used on up to five Macs. The license is personal and can be transferred between your own devices.
What do I need to run txt2mp3?
A Mac with Apple silicon is required. On first run, the app downloads its speech model, caption model, and a Python runtime, totaling about 5.5 GB.
Does my text or audio ever leave my Mac?
No, txt2mp3 operates entirely offline. Your text, recordings, and generated audio never leave your device. The app only uses the network for the initial download, license activation, and update checks.
Can I use the audio commercially?
Yes, the speech you generate is yours to use commercially. txt2mp3 does not claim ownership of the generated audio.
