Kokoro is an open-weight TTS model with 82 million parameters.
Transcribe YouTube videos to text
Convert speech to text from audio files
Generate speech from text with adjustable speed
F5-TTS & E2-TTS: Zero-Shot Voice Cloning (Unofficial Demo)
Spanish finetune for the original F5 model.
Generate speech from text with reference audio
Generate anime character speech from text
Realtime implementation of Whisper large turbo
Generate audio and SRT subtitles from text
Generate audio from text for anime characters
GPT-SoVITS for MITA!
Kokoro TTS is an open-source text-to-speech (TTS) model designed to generate high-quality audio from text. It uses an advanced neural network architecture with 82 million parameters, making it capable of producing realistic and expressive speech. The model supports multiple voices and languages, allowing users to customize the output according to their needs.
• Multi-Voice Support: Generate speech using different predefined voices.
• Customizable Parameters: Adjust pitch, tone, speed, and other aspects of the generated speech.
• Open-Source Accessibility: Freely available for use, modification, and redistribution.
• High-Quality Output: Produces natural-sounding audio with minimal robotic artifacts.
• Language Flexibility: Supports multiple languages, catering to a diverse range of users.
1. How do I install Kokoro TTS?
Kokoro TTS can be installed via its official repository. Clone the repository, install the required dependencies, and follow the setup instructions provided in the documentation.
2. Can I customize the voice and tone of the generated speech?
Yes, Kokoro TTS allows you to customize the voice, pitch, and speed of the generated speech. You can adjust these parameters through the API or command-line interface.
3. Which languages does Kokoro TTS support?
Kokoro TTS supports multiple languages, but the exact list depends on the version you are using. Refer to the official documentation for a detailed list of supported languages.