Launching today
VieNeu-TTS
Vietnamese text-to-speech with voice cloning, on-device
1 follower
Vietnamese text-to-speech with voice cloning, on-device
1 follower
VieNeu-TTS turns Vietnamese text into speech on your own computer, with instant voice cloning from a 3 to 8 second clip. It reads Vietnamese and English in one text, outputs 48 kHz audio, and installs with pip install vieneu on CPU or GPU. The v3 Turbo model is on GitHub under Apache 2.0. A newer v4 model and a paid API are at vieneu.io.
Free Options
Launch Team
AWS Activate CreditsCredits that fuel your startup’s next milestone
Promoted
Hi Product Hunt, I'm Jang. I'm hunting VieNeu-TTS, I'm not the maker. It is built by Phạm Nguyễn Ngọc Bảo in Ho Chi Minh City, and I'm hunting it because it puts Vietnamese voice cloning on a local machine.
What the project says it does:
• Runs on-device: pip install vieneu, CPU or GPU, plus a Windows installer on vieneu.io
• Instant voice cloning from a 3 to 8 second reference clip
• Vietnamese and English in the same text, 48 kHz output, 25 preset voices
• Real-time streaming, and an OpenAI-compatible API server you can run yourself with Docker
• The README reports first audio in about 115 ms on one RTX 3060 (the project's own benchmark, I haven't re-run it)
Good to know before you try it: the GitHub repo has v3 Turbo (Apache 2.0). The newer v4 is proprietary and only available through the vieneu.io API, which is paid.
Maker, if you see this: thank you for building it. Vietnamese speakers, how do the voices sound to you?