| name | free-tts |
| version | 1.0.0 |
| description | Free speech synthesis (edge-tts), no API Key or registration needed. Convert text to mp3 audio files. Triggers when user says "read aloud", "speech synthesis", "text to speech", "TTS", "generate audio", "read it", "read out", "say it", "listen" or any scenario requiring text-to-speech playback.
|
Free TTS Skill
Use edge-tts (Python library) to convert text to mp3 speech files. Under the hood calls Microsoft Edge browser online TTS API, completely free, zero registration, zero configuration, unlimited usage, direct connection in China.
Environment Dependencies
pip install edge-tts
No API Key, account, or Token needed.
Full List of Chinese Voices
| # | Voice ID | Name | Gender | Language/Style |
|---|
| 1 | zh-CN-XiaoxiaoNeural | Xiaoxiao | Female | Mandarin, Warm and Natural |
| 2 | zh-CN-XiaoyiNeural | Xiaoyi | Female | Mandarin, Lively and Playful |
| 3 | zh-CN-YunxiNeural | Yunxi | Male | Mandarin, Sunny and Energetic |
| 4 | zh-CN-YunjianNeural | Yunjian | Male | Mandarin, Passionate and Powerful |
| 5 | zh-CN-YunyangNeural | Yunyang | Male | Mandarin, Professional and Reliable |
| 6 | zh-CN-YunxiaNeural | Yunxia | Male | Mandarin, Cute and Adorable |
| 7 | zh-CN-liaoning-XiaobeiNeural | Xiaobei | Female | Dongbei Dialect, Humorous and Warm |
| 8 | zh-CN-shaanxi-XiaoniNeural | Xiaoni | Female | Shaanxi Dialect, Bright and Cheerful |
| 9 | zh-HK-HiuGaaiNeural | HiuGaai | Female | Cantonese, Friendly |
| 10 | zh-HK-HiuMaanNeural | HiuMaan | Female | Cantonese, Friendly |
| 11 | zh-HK-WanLungNeural | WanLung | Male | Cantonese, Friendly |
| 12 | zh-TW-HsiaoChenNeural | HsiaoChen | Female | Taiwan Mandarin, Friendly |
| 13 | zh-TW-HsiaoYuNeural | HsiaoYu | Female | Taiwan Mandarin, Friendly |
| 14 | zh-TW-YunJheNeural | YunJhe | Male | Taiwan Mandarin, Friendly |
Full multilingual voice list: run edge-tts --list-voices.
Usage
edge-tts --text "Hello, welcome to free speech synthesis." --write-media /var/minis/workspace/output.mp3
edge-tts --voice zh-CN-XiaoyiNeural --text "The weather is nice today" --write-media /var/minis/workspace/output.mp3
edge-tts --voice zh-CN-XiaoxiaoNeural --rate=+20% --volume=+50% \
--text "Welcome to today's news" --write-media /var/minis/workspace/output.mp3
Parameter Details
| Parameter | Description | Default |
|---|
--text | Text to synthesize (required) | - |
--write-media | Output mp3 file path (required) | - |
--voice | Voice ID | zh-CN-XiaoxiaoNeural |
--rate | Speech rate adjustment percentage | 0% |
--volume | Volume adjustment percentage | 0% |
--pitch | Pitch adjustment | 0Hz |
Output Format
Return mp3 file link for user to click and play, format:
Quick Commands
| User input | Action |
|---|
| "Read it" / "Read aloud" | Default voice (Xiaoxiao) synthesize and play |
| "Use male voice" | Switch to YunJhe (Taiwan Mandarin male) |
| "Switch" | Cycle to next voice in list |
| "Switch dialect" | Switch to Xiaoni (Shaanxi dialect) |
| "Switch to Cantonese" / "Speak Cantonese" | Switch to WanLung (Cantonese) |
| "Reset default" | Back to Xiaoxiao |
Complete Workflow
- User asks to read text aloud, or says "read it"
- Determine target voice (user specified or use current default)
- Call
edge-tts to synthesize mp3 to /var/minis/workspace/free_tts_<timestamp>.mp3
- Return audio link for user to play
Notes
- Synthesis speed: about 1-2 seconds per sentence
- Requires internet: cloud synthesis via Microsoft Edge TTS online API
- Zero cost: no account or payment needed
- Output format: 24kHz 160kbps MP3
- English and other language voices can also speak Chinese, but accent is not standard, not recommended
