edge-tts

Edge TTS Free Speech Synthesis

Generate speech using Microsoft Edge's online TTS service, free, no API Key required, supports 300+ voices.

Use Cases

User says "convert this text to speech" or "read it with a Chinese female voice"
Need to quickly generate speech without configuring an API Key
Alternative when kokoro-tts is not installed

Execution Method

Command-Line Usage

# 基本用法
edge-tts --text "你好，今天天气真不错" --write-media output.mp3

Specify Voice

edge-tts --voice zh-CN-XiaoxiaoNeural --text "你好" --write-media output.mp3

With Subtitles

edge-tts --voice zh-CN-XiaoxiaoNeural --text "你好"
--write-media output.mp3 --write-subtitles output.vtt


### Python API

```python
import edge_tts
import asyncio

async def generate():
    communicate = edge_tts.Communicate("你好，这是测试语音。", "zh-CN-XiaoxiaoNeural")
    await communicate.save("output.mp3")

asyncio.run(generate())

Recommended Chinese Voices

Voice	Gender	Style
`zh-CN-XiaoxiaoNeural`	Female	Warm and friendly (recommended)
`zh-CN-XiaoyiNeural`	Female	Lively
`zh-CN-YunjianNeural`	Male	Professional broadcast
`zh-CN-YunxiNeural`	Male	Young and natural

List all available voices

edge-tts --list-voices | grep zh-CN

Adjust Speed and Pitch

edge-tts --voice zh-CN-XiaoxiaoNeural \
  --rate "+20%" --pitch "+5Hz" \
  --text "加速播放" --write-media output.mp3

Output Specifications

By default use zh-CN-XiaoxiaoNeural(Chinese) or en-US-JennyNeural(English)
Output in MP3 format, return the file path
Requires network connection (not an offline solution)