Key Features

Low latency audio generation
Infinitely long text input handling
Small model size
CPU-based processing
Voice cloning support
Pre-made voice catalog
Python API and CLI
Local server support

The application has a small model size of 100M parameters and uses only 2 CPU cores. It can stream audio and has a faster than real-time speed of approximately 6x real-time on a MacBook Air M4. Pocket TTS also supports voice cloning and has a catalog of pre-made voices, including alba, marius, and javert, among others.


Pocket TTS can be used as a Python library, and its functionality can be accessed through a command-line interface or a local server. The application has a web interface that can be accessed at http://localhost:8000, allowing users to input text and select different voices. It also has a serve command that keeps the model in memory between requests, making it faster than the command line.

Get more likes & reach the top of search results by adding this button on your site!

Embed button preview - Light theme
Embed button preview - Dark theme
TurboType Banner
Zero to AI Engineer Program

Zero to AI Engineer

Skip the degree. Learn real-world AI skills used by AI researchers and engineers. Get certified in 8 weeks or less. No experience required.

Subscribe to the AI Search Newsletter

Get top updates in AI to your inbox every weekend. It's free!