by Ant Group
Open-weights text-to-speech models from Ant Group's Inclusion AI with voice design, cloning and tag-based control.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
Ming-3.0-tts is a family of open-weights speech generation models from Inclusion AI (Ant Group), aimed at audio stories, short-video narration and product assistants. It supports voice design from text descriptions, voice cloning, director-style instructions, and paralinguistic tags (emotion and sound effects). The site lists seven languages and two dialects, with evaluation covering Chinese, English, German and Spanish.
Two variants are listed: Ming-3.0-tts-0.5b, a dense 0.5B-parameter model for lightweight deployment, and Ming-3.0-tts-tiny, a mixture-of-experts model with 7.9B total and 1.3B active parameters that adds long-form podcast generation. A YouTube review reported a MIT license, but this was not confirmed on the fetched pages.
On Artificial Analysis, the 0.5B model scores 1069 Elo in the Controlled Voice arena (1st among open-weights models, 13th overall), 93.2% on Pronunciation Robustness (tied for 1st), and 1188 Elo in Provider Voice (3rd among open-weights, 17th overall). Weights are downloaded via Hugging Face or ModelScope and run locally; installation and cookbooks are in the repository. The developers note planned work on cross-language voice consistency, finer emotion and role control, and steadier long-form generation.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
Open text-to-speech model with natural language voice control, voice cloning, emotion and sound-effect tags, and multilingual plus Chinese dialect support. Two sizes (about 3 GB and 18 GB, the latter with multi-speaker podcasts), MIT license. Slightly ahead of Fish Audio's model on Artificial Analysis.
Closest neighbours by meaning, not just by category.
Model · Edge0
Open-weights 0.17B-parameter multilingual text-to-speech model with zero-shot voice cloning, runs on CPU via ONNX.
Model · IndexTeam
Open-weights zero-shot text-to-speech model from Bilibili's IndexTeam with voice cloning and emotion control in 5 languages.
Model · Rednote
Open-source 2B-parameter zero-shot voice-cloning text-to-speech model from Rednote's hilab team.
Model · Boson AI
Text-to-speech model from Boson AI with inline tags for emotion, style, speed, pitch, pauses and sound effects.
Model
Compact 0.6B open text-to-speech model with zero-shot voice cloning and support for 23 languages.
Model · Google
Google's Gemini 3.8 text-to-speech model for expressive, controllable speech generation via the Gemini API and AI Studio.
