Qwen-Audio-3.0-TTS-Plus
Audio modelCard summarizedA high-performance speech synthesis model, designed for high-quality speech generation scenarios
Source: qwen-audio-3.0-tts-plus model page (Alibaba Cloud Model Studio)
Where to call it
- Alibaba Cloud Model Studio
qwen-audio-3.0-tts-plus
Good for
- Audiobooks and long-form narration
- Film and video dubbing
- Brand voice and cloned spokesperson voices
Overview
Alibaba · 02Qwen-Audio, front
Qwen TTS 3.0+
A high-performance speech synthesis model, designed for high-quality speech generation scenarios
Typical price
$0.02 / 1K chars
Speech synthesis
- Max length
- —
- Languages
- System voices: Mandarin, English. Cloned voices: Chinese (Mandarin, Cantonese and 19 regional dialects), English, Japanese, Korean, Russian, French, German, Portuguese, Thai, Indonesian, Vietnamese, Spanish, Italian, Malay, Filipino, Arabic
- Modes
- text-to-speech, voice cloning
- Released
- 14 Jul 2026
The quality-first model of Qwen-Audio-3.0 TTS. Compared with the previous version it adds low-resource languages and Chinese dialects, follows free-form style instructions and fine-grained tags for emotion, tone, character, rate and volume, and is more robust to noisy reference audio. Aimed at audiobooks, dubbing, brand voices and premium voice services. model page
qwen-audio-3.0-tts-plus- Typical price
- $0.02 / 1K chars
- Languages
- System voices: Mandarin, English. Cloned voices: Chinese (Mandarin, Cantonese and 19 regional dialects), English, Japanese, Korean, Russian, French, German, Portuguese, Thai, Indonesian, Vietnamese, Spanish, Italian, Malay, Filipino, Arabic
- Weights
- Closed
- Released
- 14 Jul 2026
- text-to-speech
- voice cloning
- instruction control (emotion, tone, rate, volume, style)
- fine-grained tags
- streaming over WebSocket