Coqui TTS — синтез и клонирование голоса

★ 33k

Coqui TTS is an open-source speech synthesis library: it turns written text into a spoken voice. Mind the direction — this is text-to-speech, not transcription; to go the other way and get text out of a recording you need speech recognition tools such as Whisper. Reach for it when you need to voice a video, an article or a newsletter, give a bot or assistant a voice, narrate a course without a microphone and a studio, or clone your own voice from a short sample and have it read any text afterwards. It ships more than 1,100 pretrained models across many languages; the flagship XTTSv2 covers 16 languages, supports voice cloning and streams audio with under 200 ms latency, which is fast enough for live conversation. It integrates with Bark and Tortoise and includes tooling to fine-tune on your own recordings. Everything runs locally, so your audio never leaves your machine.