AlibabaQwen Audio
Qwen Audio 3.1 TTS
Qwen Audio 3.1 TTS offers Plus for detailed voice performance and Flash for low-latency interaction. Both accept text and generate speech.
Practical use
Practical guidance. Choose the variant from the speech workflow. For live replies, measure delay to the first audible response and interruption handling. For recorded narration, review names, numbers and emotional transitions across repeated takes. Save delivery instructions with the script.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API |
| Model identifiers | qwen-audio-3.1-tts-plusqwen-audio-3.1-tts-flash |
| Controls | Natural-language instructions and tags control tone, emotion, speed, persona and volume.Qwen Audio 3.1 TTS Flash |
| Access | Beijing and Singapore APIs; streaming and HTTP interfaces.Qwen Audio 3.1 TTS Plus |
| Voice reference | Voice cloning accepts reference recordings and is designed to tolerate noise and reverberation.Qwen Audio 3.1 TTS Plus |
Related models
- Qwen Audio 3.1 Realtime
Alibaba · Qwen Audio
Full-duplex speech and text model family with Plus and Flash variants and acoustic, semantic and manual turn controls.
Launch date not confirmedAPI - Qwen Audio 3.1 TTS Next
Alibaba · Qwen Audio
Chinese and English audio generation model that combines speech, sound effects and ambient audio.
Launch date not confirmedAPI