The Kokoro-82M model provides open-weight text-to-speech with 82 million parameters.
Kokoro
Kokoro provides an inference library for its open-weight 82-million-parameter text-to-speech model.
About Kokoro
Kokoro is an open-weight text-to-speech model with 82 million parameters, distributed with a Python inference library on GitHub. The repository, hexgrad/kokoro, can be installed with pip and includes code, examples, a demo, and tests. Its weights are described as Apache-licensed for use in production environments and personal projects.
You can use the KPipeline class to convert written text into audio files, select a voice such as af_heart, and set a language code for the input. The documentation shows running the pipeline in Google Colab, installing espeak-ng, and saving output as WAV files at 24,000 Hz. Under the hood, the library uses misaki for grapheme-to-phoneme conversion.
It suits developers and researchers comfortable with Python who want to run text-to-speech locally or in their own applications. Support covers multiple language codes, including American and British English, Spanish, French, Hindi, Italian, Japanese, Brazilian Portuguese, and Mandarin Chinese.
Key features
- Pip-installable Python library for Kokoro-82M inference
- KPipeline interface to convert text to WAV audio
- Language codes for English, Spanish, French, Hindi, and more
- Uses misaki library for grapheme-to-phoneme conversion
- Uses espeak-ng for English fallback and some languages
Best for: Developers and researchers who want to run an open-weight text-to-speech model in Python projects.
Good for
- Open-weight 82M-parameter model with Apache-licensed weights
- Pip install with documented Colab usage examples
- Multiple language codes and selectable voices
Watch out for
- Requires Python, pip, and additional system installs
- Some languages need extra misaki installs
Kokoro at a glance
- Website
- github.com/hexgrad/kokoro
- Category
- AI & Machine Learning › AI Voice & Audio
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 5, 2026 · Online when checked
- Listed
- Oct 5, 2026
Kokoro alternatives
Tools and sites you could use instead, ranked by how closely they match.
The KittenTTS project offers a text-to-speech model described as under 25 MB in size.
Open source
24 alternatives
The whisper.cpp project ports OpenAI’s Whisper speech recognition model to C and C++.
Open source
24 alternatives
Generate audio from text prompts with Bark, a model for text-prompted audio creation.
Open source
16 alternatives
RVC WebUI is a framework for training voice conversion models with short voice recordings.
Open source
24 alternatives
Tortoise is a multi-voice text-to-speech system trained with an emphasis on quality.
Open source
24 alternativesPopular in AI Voice & Audio
Well-known listings from the same topic.
Google Illuminateilluminate.google.comAlteredaltered.aiELSA Speakelsaspeak.comKreadoAIkreadoai.comPodwisepodwise.aiAddsubtitleaddsubtitle.ai
Kokoro FAQ
What is Kokoro?
Kokoro is an inference library for Kokoro-82M, an open-weight 82-million-parameter text-to-speech model.
How do you use Kokoro?
You install the kokoro package with pip, initialize KPipeline with a language code, and generate audio from text. The documentation shows running it in Google Colab.
Which languages does Kokoro support?
Language codes include American and British English, Spanish, French, Hindi, Italian, Japanese, Brazilian Portuguese, and Mandarin Chinese.
Is Kokoro open source?
Yes, the code is shared on GitHub and the weights are described as Apache-licensed.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.