Pocket TTS
Pocket TTS is a lightweight text-to-speech application designed to run on CPUs.
What Pocket TTS is
Pocket TTS is a lightweight text-to-speech application designed to run on CPUs. Its 100-million-parameter model supports streaming audio, with the first audio chunk generated in about 200 milliseconds. The documentation reports about six times real-time speed on a MacBook Air M4 CPU using two cores. It handles long text inputs and supports six languages, with WAV export. Interfaces include a Python API, command-line tool, and local HTTP server; a browser demo can generate speech without installation, though browser implementations are not officially supported. Installation is documented for Linux, macOS, and Windows. The package supports Python 3.10 through 3.14 and requires PyTorch 2.5 or newer, but not the GPU version. On Linux, the default PyTorch installation may download several gigabytes of CUDA runtime wheels unless users select the CPU-only index. Pocket TTS is free and allows commercial use. Voice cloning is subject to restrictions against impersonation without explicit lawful consent and other illegal, harmful, deceptive, or privacy-invasive uses. Model files are listed under a CC-BY-4.0 license, and access requires agreeing to share contact information.
Good
- Runs on CPUs and supports streaming audio.
- Provides Python API, command-line, and HTTP interfaces.
- Supports six languages and WAV export.
- Commercial use is permitted.
Watch out
- Requires Python 3.10 through 3.14 and PyTorch 2.5 or newer.
- Browser implementations are not officially supported.
- Linux defaults may download large CUDA runtime wheels.
- Model access requires sharing contact information.
Who it suits
It suits developers and users who want CPU-based speech generation through a Python API, CLI, or local server. It may also suit commercial projects, provided use follows the stated restrictions.
Verdict
Pocket TTS offers CPU-oriented speech generation with several interfaces and a browser demo. Users should check the runtime requirements, model access condition, and voice-cloning restrictions before adopting it.
Plans and prices
From Pocket TTS's own pricing page, checked 12 Oct 2026.
| Plan | Price | What it gives you |
|---|---|---|
| Pocket TTS | Free | Open source · CPU-based text-to-speech · Python API and CLI |
The facts
Each one read from Pocket TTS's own pages, with the date it was checked.
What it does
- API access
- Yeschecked 20 Sep 2026
- Commercial use
- Yeschecked 20 Sep 2026
- CPU performance
- The project reports about six times real-time generation on a MacBook Air M4 CPU using two CPU cores.checked 12 Oct 2026
- Export formats
- WAVchecked 20 Sep 2026
- Integrations
- The project lists community projects including a Wyoming protocol Docker container for Home Assistant Voice and a Unity 6 integration.checked 12 Oct 2026
- Interfaces
- It offers a Python library, command-line interface, and local HTTP server with a web interface at localhost:8000.checked 12 Oct 2026
- Languages
- The documented pretrained language models cover English, French, German, Portuguese, Italian, and Spanish.checked 12 Oct 2026
- License and access
- The Hugging Face model card lists the model license as CC-BY-4.0 and says users must agree to share contact information to access the model files.checked 11 Oct 2026
- Long inputs
- Pocket TTS can handle infinitely long text inputs.checked 12 Oct 2026
- Long text
- The model can handle infinitely long text inputs.checked 11 Oct 2026
- Model size
- The model has 100 million parameters.checked 12 Oct 2026
- Purpose
- Pocket TTS is a lightweight text-to-speech application designed to run efficiently on CPUs.checked 11 Oct 2026
- Requirements
- The package supports Python 3.10 through 3.14 and requires PyTorch 2.5 or later.checked 11 Oct 2026
- Runtime requirements
- The package supports Python 3.10 through 3.14 and requires PyTorch 2.5 or newer; it does not require the GPU version of PyTorch.checked 12 Oct 2026
- Security and use policy
- The project prohibits voice impersonation or cloning without explicit and lawful consent, along with deceptive, harmful, or privacy-invasive uses.checked 12 Oct 2026
- Streaming and latency
- It supports audio streaming and reports about 200 ms to produce the first audio chunk.checked 12 Oct 2026
- Support
- The maker invites users to open issues or pull requests on GitHub and accepts contributions.checked 11 Oct 2026
- Support and contributions
- The project invites users to open GitHub issues or pull requests and links to development instructions.checked 12 Oct 2026
Where it runs
- Browser availability
- The maker offers a browser demo and says the model can run client-side in a browser, while noting that browser implementations are not officially supported.checked 11 Oct 2026
- Browser demo
- The documentation links to a Kyutai website demo that can generate speech in the browser without installation.checked 12 Oct 2026
- Platforms
- windows,macos,linux,api,self_hostedchecked 20 Sep 2026
- Platform setup
- The maker documents installation on Linux, macOS, and Windows; on Linux, its default PyTorch installation may download several gigabytes of CUDA runtime wheels unless users choose the CPU-only index.checked 11 Oct 2026
Money
- Free plan
- Yeschecked 20 Sep 2026
The company
- Headquarters
- Paris, Francechecked 28 Sep 2026
Alternatives to Pocket TTS
- 1
iSpeech
Free planOpen sourceAndroidiOSWebAPI8.5Free plan - 2
Voicebox
Free planOpen sourceAndroidiOSWebAPI8.5From $12 a year - 3
Resemble AI
Free planAndroidiOSWebAPI8.3From $280 a month - 4
Speechify
Free planAndroidiOSWebAPI8.3From $10 a month - 5
Voice.ai
Free planAndroidiOSWebAPI8.3From $5 a month - 6
Fliki
Free planAndroidiOSWebAPI8.2From $28 a month