Tasks and inputs
Qwen-Audio-3.1-TTS Next is listed for Audio, Audio generation, Text-to-speech. It accepts text, audio and produces audio. Use the publisher’s input format and preprocessing for your chosen deployment.
SyncDev model directory · Profile updated Nov 17, 2026
Qwen-Audio-3.1-TTS Next is Alibaba's unified audio generator: speech, multi-speaker dialogue, effects and ambience in one pass from text and timestamps.
The specification score summarizes recorded features and limits. It is not a benchmark result.
Pricing conditions: No published price found on Model Studio pages checked; confirm in the Model Studio console for your region.
Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.
Accepts
Produces
What if the voice track, the footsteps and the rain outside the window came out of one model call?
That is the premise of qwen-audio-3.1-tts-next. Despite the TTS in its name, Alibaba classes it as audio generation and describes it as complete sound-scene generation. Instead of producing a voice you then mix, it composes the whole clip in a single pass.
Output is audio.
Alibaba lists a wide spread of outputs: multilingual speech, single-speaker narration, multi-speaker dialogue, podcasts, cinematic soundscapes, ambient beds and sound effects. The strengths it claims are expressive speech, a voice identity that stays consistent across a long piece, flexible timing control and soundscape generation.
For a podcast producer that might mean two hosts, an intro sting and room tone from one request. For a game studio, a scene of barks over battlefield ambience.
We would rather list the gaps than paper over them.
| Item | Status |
|---|---|
| Release date | 22 October 2026 in the international changelog; 21 September 2026 in the China (Beijing) section |
| Documentation | Not covered by the Qwen TTS guide, which documents only Flash and Plus |
| Request schema | Not published on the pages we read |
| Price | Not found |
| Output formats, maximum duration | Not found |
The two dates likely reflect different regional rollouts, with Beijing first. If you are on the international service, 22 October is the date that applies to you. Because the dedicated guide does not cover this model, do not assume that the Flash/Plus tag set or instruction field behaves the same way here. Ask support or inspect the console's sample request.
Qwen-Audio-3.1-Realtime-Plus holds conversations. The Flash and Plus TTS models read text aloud, fast or expressively. Next is the only one in the line aimed at producing finished sound design. If all you need is a narrator, use one of the TTS models; you will pay for, and debug, less.
Generation quality on soundscapes is subjective, so make the test concrete.
Then compare the result with your current workflow of separate TTS plus a stock library. The question is not whether it beats a sound designer, but whether it saves enough editing time to be worth adopting.
Related models sit under audio generation and on the Alibaba provider page.
Compare the model on the work you need to complete.
Qwen-Audio-3.1-TTS Next is listed for Audio, Audio generation, Text-to-speech. It accepts text, audio and produces audio. Use the publisher’s input format and preprocessing for your chosen deployment.
Compare pronunciation, voice consistency and time to the first playable audio. Check the service’s billing unit for the actual script.
Use the model identifier and account requirements for the selected API route. Confirm its request limits and output format with one complete task before connecting a larger workflow.
These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.
ElevenLabs
SFX v3 is ElevenLabs' newest text-to-sound-effects model, better at multi-part prompts, music and vocals, with seamless looping and less clipping.
StepFun
StepAudio 3 Gen Preview creates a complete audio scene from text, audio references and role descriptions.