Kokoro TTS

A CLI text-to-speech tool using the Kokoro model, supporting multiple languages, voices (with blending), and various input formats including EPUB books and PDF documents.

Features

• Multiple language and voice support
• Voice blending with customizable weights
• EPUB, PDF and TXT file input support
• Standard input (stdin) and | piping from other programs
• Streaming audio playback
• Split output into chapters
• Adjustable speech speed
• WAV and MP3 output formats
• Chapter merging capability
• Detailed debug output option
• GPU Support

Demo

Kokoro TTS is an open-source CLI tool that delivers high-quality text-to-speech right from your terminal. Think of it as your personal voice studio, capable of transforming any text into natural-sounding speech with minimal effort.

https://github.com/user-attachments/assets/8413e640-59e9-490e-861d-49187e967526

Demo Audio (MP3) | Demo Audio (WAV)

TODO

• Add GPU support
• Add PDF support
• Add GUI

Prerequisites

• Python 3.11-3.12 (Python 3.13+ is not currently supported)

Installation

Method 1: Install from PyPI (Recommended)

The easiest way to install Kokoro TTS is from PyPI:

# Using uv (recommended)
uv tool install kokoro-tts

# Using pip
pip install kokoro-tts

After installation, you can run:

kokoro-tts --help

Method 2: Install from Git

Install directly from the repository:

# Using uv (recommended)
uv tool install git+https://github.com/nazdridoy/kokoro-tts

# Using pip
pip install git+https://github.com/nazdridoy/kokoro-tts

Method 3: Clone and Install Locally

1. Clone the repository:

git clone https://github.com/nazdridoy/kokoro-tts.git
cd kokoro-tts

2. Install the package:

With uv (recommended):

uv venv
uv pip install -e .

With pip:

python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
pip install -e .

3. Run the tool:

# If using uv
uv run kokoro-tts --help

# If using pip with activated venv
kokoro-tts --help

Method 4: Run Without Installation

If you prefer to run without installing:

1. Clone the repository:

git clone https://github.com/nazdridoy/kokoro-tts.git
cd kokoro-tts

2. Install dependencies only:

With uv:

uv venv
uv sync

With pip:

python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
pip install -r requirements.txt

3. Run directly:

# With uv
uv run -m kokoro_tts --help

# With pip (venv activated)
python -m kokoro_tts --help

Download Model Files

After installation, download the required model files to your working directory:

# Download voice data (bin format is preferred)
wget https://github.com/nazdridoy/kokoro-tts/releases/download/v1.0.0/voices-v1.0.bin

# Download the model
wget https://github.com/nazdridoy/kokoro-tts/releases/download/v1.0.0/kokoro-v1.0.onnx

The script requires voices-v1.0.bin and kokoro-v1.0.onnx to be present in the same directory where you run the kokoro-tts command.

Supported voices:

Usage

Basic Usage

kokoro-tts <input_text_file> [<output_audio_file>] [options]

Commands

• -h, --help: Show help message
• --help-languages: List supported languages
• --help-voices: List available voices
• --merge-chunks: Merge existing chunks into chapter files

Options

• --stream: Stream audio instead of saving to file
• --speed <float>: Set speech speed (default: 1.0)
• --lang <str>: Set language (default: en-us)
• --voice <str>: Set voice or blend voices (default: interactive selection)
  • Single voice: Use voice name (e.g., "af_sarah")
  • Blended voices: Use "voice1:weight,voice2:weight" format
• --split-output <dir>: Save each chunk as separate file in directory
• --format <str>: Audio format: wav or mp3 (default: wav)
• --debug: Show detailed debug information during processing

Input Formats

• .txt: Text file input
• .epub: EPUB book input (will process chapters)
• .pdf: PDF document input (extracts chapters from TOC or content)
• - or /dev/stdin (Linux/macOS) or CONIN$ (Windows): Standard input (stdin)

Examples

# Basic usage with output file
kokoro-tts input.txt output.wav --speed 1.2 --lang en-us --voice af_sarah

# Read from standard input (stdin)
echo "Hello World" | kokoro-tts - --stream
cat input.txt | kokoro-tts - output.wav

# Cross-platform stdin support:
# Linux/macOS: echo "text" | kokoro-tts - --stream
# Windows: echo "text" | kokoro-tts - --stream
# All platforms also support: kokoro-tts /dev/stdin --stream (Linux/macOS) or kokoro-tts CONIN$ --stream (Windows)

# Use voice blending (60-40 mix)
kokoro-tts input.txt output.wav --voice "af_sarah:60,am_adam:40"

# Use equal voice blend (50-50)
kokoro-tts input.txt --stream --voice "am_adam,af_sarah"

# Process EPUB and split into chunks
kokoro-tts input.epub --split-output ./chunks/ --format mp3

# Stream audio directly
kokoro-tts input.txt --stream --speed 0.8

# Merge existing chunks
kokoro-tts --merge-chunks --split-output ./chunks/ --format wav

# Process EPUB with detailed debug output
kokoro-tts input.epub --split-output ./chunks/ --debug

# Process PDF and split into chapters
kokoro-tts input.pdf --split-output ./chunks/ --format mp3

# List available voices
kokoro-tts --help-voices

# List supported languages
kokoro-tts --help-languages

Features in Detail

EPUB Processing

• Automatically extracts chapters from EPUB files
• Preserves chapter titles and structure
• Creates organized output for each chapter
• Detailed debug output available for troubleshooting

Audio Processing

• Chunks long text into manageable segments
• Supports streaming for immediate playback
• Voice blending with customizable mix ratios
• Progress indicators for long processes
• Handles interruptions gracefully

Output Options

• Single file output
• Split output with chapter organization
• Chunk merging capability
• Multiple audio format support

Debug Mode

• Shows detailed information about file processing
• Displays NCX parsing details for EPUB files
• Lists all found chapters and their metadata
• Helps troubleshoot processing issues

Input Options

• Text file input (.txt)
• EPUB book input (.epub)
• Standard input (stdin)
• Supports piping from other programs

Contributing

This is a personal project. But if you want to contribute, please feel free to submit a Pull Request.

License

This project is licensed under the MIT License. See the LICENSE file for details.

Acknowledgments

• Kokoro-ONNX