OpenAI TTS Skill

Text-to-speech conversion using OpenAI's TTS API for generating high-quality, natural-sounding audio from text.

Installation

pip install openai pydub

# For audio processing (pydub dependency)
# macOS
brew install ffmpeg

# Ubuntu/Debian
sudo apt install ffmpeg

Setup

Set your OpenAI API key:

export OPENAI_API_KEY="sk-..."

Usage

Command Line

# Basic usage
python openai_tts.py "Hello world" -o output.mp3

# From file
python openai_tts.py -f article.txt -o article.mp3

# With voice selection
python openai_tts.py "Your text" -o output.mp3 --voice nova

# High quality
python openai_tts.py "Your text" -o output.mp3 --model tts-1-hd

# Adjust speed (0.25 to 4.0)
python openai_tts.py "Your text" -o output.mp3 --speed 1.5

# Pipe input
echo "Hello world" | python openai_tts.py -o output.mp3

# Verbose mode
python openai_tts.py "Test" -o test.mp3 -v

# List available voices
python openai_tts.py --list-voices

As Module

from openai_tts import generate_tts

# Basic
generate_tts("Hello world", "output.mp3")

# With options
generate_tts(
    text="Your text here",
    output_path="output.mp3",
    voice="nova",
    model="tts-1-hd",
    response_format="mp3",
    speed=1.25,  # 0.25 to 4.0
    verbose=True
)

Voices

Models

Features

• Auto-chunking: Automatically splits text longer than 4096 characters
• Multiple formats: mp3, opus, aac, flac
• 6 voices: Male and female options
• Pipe support: Read from stdin

Output Formats

License

MIT