Qwen3-TTS

There are 3 models:

1. VoiceDesign: Creates voices from text descriptions of what you want the voice to sound like
2. CustomVoice: Lets you control voice style through instructions; includes 9 pre-made premium voices with different genders, ages, languages, and dialects
3. Base: Core model that can clone any voice from just 3 seconds of audio; can also be fine-tuned for specialized use cases

All three models support 10 languages (Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish, Italian) and appear to have the same feature availability.

My personal favorite feature is "voice design", where you can actually prompt the speech style, it's a game changer!