generate-voice

Generate voice audio from text using free Microsoft TTS — no API key needed.

400+ voices in 100+ languages. One command. Free and unlimited.

Quick Start

npx generate-voice "Hello, how are you?"

That's it. You get an .mp3 file.

Prerequisites

• Node.js >= 14
• Python 3 with pip (edge-tts is auto-installed on first run)

Usage

Generate speech

# Default voice (en-US-GuyNeural)
npx generate-voice "Hello world"

# Choose a voice
npx generate-voice "Bonjour le monde" --voice fr-FR-HenriNeural

# Save to specific file
npx generate-voice "Hello" --output greeting.mp3

# Adjust speed
npx generate-voice "This is fast" --rate "+30%"
npx generate-voice "This is slow" --rate "-20%"

# Adjust pitch
npx generate-voice "Higher pitch" --pitch "+5Hz"
npx generate-voice "Lower pitch" --pitch "-5Hz"

# Combine options
npx generate-voice "Custom voice" --voice en-GB-SoniaNeural --rate "+10%" --output custom.mp3

Browse voices

# List all 400+ voices
npx generate-voice --voices

# Filter by language
npx generate-voice --voices --lang en    # English
npx generate-voice --voices --lang fr    # French
npx generate-voice --voices --lang de    # German
npx generate-voice --voices --lang ja    # Japanese
npx generate-voice --voices --lang es    # Spanish

Setup (optional)

npx generate-voice setup

Checks Python 3 and installs edge-tts. This runs automatically on first use, but you can run it manually to verify your environment.

Options

| Flag | Description | Default | |------|-------------|---------| | -v, --voice <name> | TTS voice name | en-US-GuyNeural | | -o, --output <file> | Output file path | Auto from text | | -r, --rate <rate> | Speech rate (+20%, -10%) | Normal | | -p, --pitch <pitch> | Voice pitch (+5Hz, -2Hz) | Normal | | --voices | List all available voices | — | | -l, --lang <code> | Filter voice list by language | — |

Popular Voices

| Voice | Language | Gender | |-------|----------|--------| | en-US-GuyNeural | English (US) | Male | | en-US-JennyNeural | English (US) | Female | | en-US-AriaNeural | English (US) | Female | | en-GB-SoniaNeural | English (UK) | Female | | en-GB-RyanNeural | English (UK) | Male | | fr-FR-HenriNeural | French | Male | | de-DE-ConradNeural | German | Male | | es-ES-AlvaroNeural | Spanish | Male | | ja-JP-KeitaNeural | Japanese | Male | | zh-CN-XiaoxiaoNeural | Chinese | Female |

How It Works

Uses edge-tts, a Python library that accesses Microsoft Edge's free text-to-speech API. No API key, no account, no limits.

License

MIT