Reelina

Qwen-Audio 3.1 TTS Flash

Alibaba · Sound · 1 coins per generation, shown before you run it

Text to synthesize into speech audio. Tuned for low-latency interaction; supports multiple foreign languages and Chinese dialects, and free-style instruction control. Supports SSML and LaTeX input when enable_ssml is true.

System voice for qwen-audio-3.1-tts-flash (Singapore deployment). Lists the 70 Singapore-available system voices (4 multilingual/dialect + 15 premium English + 51 secondary-language). Premium-Chinese and other-Mandarin voices from the official list are Beijing-region only and are excluded. Secondary-language voices use the fully qualified id form qwen-audio-3.1-tts-flash-<name>_<lang>. The field also accepts any voice id outside this list and forwards it as-is. Dialect/language is selected at request time via the instruction field (e.g. 请讲上海话).

Output audio format.

Output sample rate in Hz.

Speech rate multiplier.

Pitch multiplier.

Output volume.

Output bitrate in kbps. Applies to opus only.

Language hint list, for example ["zh"]. Supported values: zh, en, ja, ko, fr, de, pt, it, vi, id.

Free-style instruction controlling emotion, tone, role, dialect, or speaking style, for example "请讲上海话".

enable ssml

Treat text as SSML markup.

About Qwen-Audio 3.1 TTS Flash

Alibaba's low-latency 3.1-generation text-to-speech model: rich system voices covering multiple foreign languages and Chinese dialects (Shanghainese, Cantonese, Northeastern, Sichuanese, and more), with free-style instruction control over emotion, tone, role, speed, and volume. Tuned for first-packet latency in real-time interaction.

Specifications

ProviderQwen (Alibaba)
TypeAudio generation
Cost per run 1 coins
InputText prompt

Alternatives to Qwen-Audio 3.1 TTS Flash

Models in this workspace that do the same job, closest in price first. Your coin balance works across all of them.

Frequently asked questions

What is Qwen-Audio 3.1 TTS Flash?
Alibaba's low-latency 3.1-generation text-to-speech model: rich system voices covering multiple foreign languages and Chinese dialects (Shanghainese, Cantonese, Northeastern, Sichuanese, and more), with free-style instruction control over emotion, tone, role, speed, and volume. Tuned for first-packet latency in real-time interaction.
How much does Qwen-Audio 3.1 TTS Flash cost?
Qwen-Audio 3.1 TTS Flash costs 1 coins per run, deducted from your Reelina balance. The exact cost is shown before you press Generate, and one balance covers every model — there is no separate subscription per model.
What do I need to use Qwen-Audio 3.1 TTS Flash?
A text prompt describing what you want. No upload needed.
Can I use the results commercially?
Your generations are yours to use, subject to our terms and acceptable use policy. For anything business-critical, check the licence of the specific model you pick.
How long are my files kept?
Uploads are used only to produce your result. Creations are private by default and everything is deleted one week after it is generated.

Looking for something else? Browse every model.