Kokoro is an open-weight TTS model with 82 million parameters.

Generate realistic audio from text

Transcribe audio from microphone, file, or YouTube link

Convert spoken words into text

Transcribe or translate audio and YouTube videos

High-fidelity Text-To-Speech

Turn text into speech with customizable voice, rate, and pitch

Request evaluation of a speech recognition model

Efficient, fast, and natural text to speech with StyleTTS 2!

Generate realistic voices from text

Generate anime character speech from text

Generate speech from text

Fast, efficient, & multilingual text-to-speech

Spanish finetune for the original F5 model.

Transcribe voice to text

Convert text to speech with Next-gen Kaldi

Realtime implementation of Whisper large turbo

Text to Audio (Sound SFX) Generator

ExpressivText-to-Speech

Generate audio from text or modify voice pitch