> ## Documentation Index > Fetch the complete documentation index at: https://docs.shuyou.ai/llms.txt > Use this file to discover all available pages before exploring further. # Qwen3 TTS Flash > Create asynchronous text-to-speech tasks with qwen3-tts-flash via POST /v1/predictions. ## Voice presets (`input.voice`) Pass a **voice preset name** in `input.voice`. The table below lists Qwen TTS voices supported by **`qwen3-tts-flash`** (and related Qwen TTS model families). | `voice` | Name | Description | Languages | | ------------- | ------ | ----------------------------------------------------------- | ------------------------------------------------------------------------------------------ | | `Cherry` | 芊悦 | Sunny, upbeat, natural female | Mandarin, English, French, German, Russian, Italian, Spanish, Portuguese, Japanese, Korean | | `Serena` | 苏瑶 | Gentle female | Same as above | | `Ethan` | 晨煦 | Warm, energetic male (Mandarin with slight northern accent) | Same as above | | `Chelsie` | 千雪 | Anime-style virtual girlfriend (female) | Same as above | | `Momo` | 茉兔 | Playful, teasing female | Same as above | | `Vivian` | 十三 | Cute but feisty female | Same as above | | `Moon` | 月白 | Cool, handsome male | Same as above | | `Maia` | 四月 | Intellectual and gentle female | Same as above | | `Kai` | 凯 | Soothing male | Same as above | | `Nofish` | 不吃鱼 | Designer male (non-retroflex accent) | Same as above | | `Bella` | 萌宝 | Cute young female | Same as above | | `Jennifer` | 詹妮弗 | Cinematic American female | Same as above | | `Ryan` | 甜茶 | Dramatic, expressive male | Same as above | | `Katerina` | 卡捷琳娜 | Mature female | Same as above | | `Aiden` | 艾登 | American male | Same as above | | `Eldric Sage` | 沧明子 | Wise elder male | Same as above | | `Mia` | 乖小妹 | Obedient, gentle female | Same as above | | `Mochi` | 沙小弥 | Precocious young male | Same as above | | `Bellona` | 燕铮莺 | Powerful, clear female (period drama style) | Same as above | | `Vincent` | 田叔 | Husky, gritty male | Same as above | | `Bunny` | 萌小姬 | Very cute young female | Same as above | | `Neil` | 阿闻 | Professional news anchor male | Same as above | | `Elias` | 墨讲师 | Educational female narrator | Same as above | | `Arthur` | 徐大爷 | Folksy storyteller male | Same as above | | `Nini` | 邻家妹妹 | Sweet younger-sister female | Same as above | | `Seren` | 小婉 | Calm, sleep-aid female | Same as above | | `Pip` | 顽屁小孩 | Mischievous boy | Same as above | | `Stella` | 少女阿月 | Sweet magical-girl female | Same as above | | `Bodega` | 博德加 | Warm Spanish uncle (male) | Same as above | | `Sonrisa` | 索尼莎 | Cheerful Latina female | Same as above | | `Alek` | 阿列克 | Russian male | Same as above | | `Dolce` | 多尔切 | Laid-back Italian uncle (male) | Same as above | | `Sohee` | 素熙 | Warm Korean female | Same as above | | `Ono Anna` | 小野杏 | Feisty childhood-friend female | Same as above | | `Lenn` | 莱恩 | Rational German youth (male) | Same as above | | `Emilien` | 埃米尔安 | Romantic French male | Same as above | | `Andre` | 安德雷 | Magnetic, steady male | Same as above | | `Radio Gol` | 拉迪奥·戈尔 | Sports commentator male | Same as above | | `Jada` | 上海-阿珍 | Shanghai dialect female | Shanghai dialect + same multilingual set | | `Dylan` | 北京-晓东 | Beijing dialect male | Beijing dialect + same multilingual set | | `Li` | 南京-老李 | Nanjing dialect male | Nanjing dialect + same multilingual set | | `Marcus` | 陕西-秦川 | Shaanxi dialect male | Shaanxi dialect + same multilingual set | | `Roy` | 闽南-阿杰 | Taiwanese Hokkien male | Hokkien + same multilingual set | | `Peter` | 天津-李彼得 | Tianjin dialect male (comic duo style) | Tianjin dialect + same multilingual set | | `Sunny` | 四川-晴儿 | Sichuan dialect female | Sichuan dialect + same multilingual set | | `Eric` | 四川-程川 | Sichuan dialect male | Sichuan dialect + same multilingual set | | `Rocky` | 粤语-阿强 | Cantonese male | Cantonese + same multilingual set | | `Kiki` | 粤语-阿清 | Cantonese female | Cantonese + same multilingual set | ## Language (`input.language`) Optional. Sets the synthesis language (Qwen API field: `language_type`). Default **`Auto`**. | Value | When to use | | ------------ | ------------------------------------------------------------------------------------------------------------------- | | `Auto` | Mixed-language text or uncertain language. The model matches pronunciation per segment; accuracy is not guaranteed. | | `Chinese` | Single-language Chinese text — typically better quality than `Auto` | | `English` | Single-language English text | | `German` | Single-language German text | | `Italian` | Single-language Italian text | | `Portuguese` | Single-language Portuguese text | | `Spanish` | Single-language Spanish text | | `Japanese` | Single-language Japanese text | | `Korean` | Single-language Korean text | | `French` | Single-language French text | | `Russian` | Single-language Russian text | For monolingual content, specify the language explicitly instead of `Auto` for best results. ## OpenAPI ````yaml en/api-reference/audio-series/qwen/qwen3-tts-flash-generate.json POST /v1/predictions openapi: 3.1.0 info: title: Qwen3 TTS Flash version: 1.0.0 servers: - url: https://api.shuyou.ai security: - bearerAuth: [] paths: /v1/predictions: post: tags: - Audio Generation summary: Qwen3 TTS Flash description: >- Qwen3 TTS Flash (`qwen3-tts-flash`) converts text to speech asynchronously via POST /v1/predictions. Poll [Get a prediction](/en/api-reference/task-management/get-prediction) using `data.task_id`, or configure top-level `webhook` for completion callbacks. operationId: createQwen3TtsFlashPrediction requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/Qwen3TtsFlashPredictionRequest' examples: text_to_speech: summary: Text to speech value: model: qwen3-tts-flash input: prompt: >- At the jet bridge, Leo took his first step onto the metal walkway. He could smell faint fumes from the runway. The plane loomed ahead like a quiet animal. As he approached the door, he paused, gave the flight attendant a small smile, and entered the long, low-lit cabin.Engines roared. Leo grasped the armrest as the plane gathered speed and lifted into the sky. The city below shrank into a dim grid of lights. At cruising altitude, the engines settled into a low hum. The atmosphere outside felt thin and silent. voice: Cherry language: Auto webhook: https://api.shuyou.ai/backend/api/callback text_to_speech_minimal: summary: Minimal request value: model: qwen3-tts-flash input: prompt: Welcome to ShuYou AI documentation. voice: Cherry language: Auto responses: '200': description: Async task created content: application/json: schema: $ref: '#/components/schemas/PredictionTaskResponse' example: data: task_id: 2c4d50261173430290971a2395a3b607 task_status: processing '400': description: Invalid request content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '401': description: Unauthorized content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '402': description: Insufficient quota content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '403': description: Model access denied content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '429': description: Rate limit exceeded content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '500': description: Internal server error content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' components: schemas: Qwen3TtsFlashPredictionRequest: type: object required: - model - input properties: model: type: string description: Model ID. Use `qwen3-tts-flash` for this endpoint. enum: - qwen3-tts-flash default: qwen3-tts-flash example: qwen3-tts-flash input: $ref: '#/components/schemas/Qwen3TtsFlashInput' webhook: type: string format: uri description: >- Optional HTTPS callback URL when the task completes, fails, or is cancelled. PredictionTaskResponse: type: object properties: data: type: object properties: task_id: type: string description: Async task ID. Use with Get a prediction or webhook correlation. example: 2c4d50261173430290971a2395a3b607 task_status: type: string description: Current task status, e.g. `processing`. example: processing required: - task_id - task_status required: - data ErrorResponse: type: object properties: error: type: object properties: code: type: string message: type: string type: type: string Qwen3TtsFlashInput: type: object required: - prompt properties: prompt: type: string description: Text content to synthesize into speech. example: Welcome to ShuYou AI documentation. voice: type: string description: >- Qwen TTS voice preset. Pass the preset name exactly as listed (e.g. `Cherry`, `Serena`, `Eldric Sage`, `Ono Anna`). See the voice table on this page for names, descriptions, and supported languages. All listed presets are supported by `qwen3-tts-flash` unless noted otherwise on the model page. example: Cherry enum: - Cherry - Serena - Ethan - Chelsie - Momo - Vivian - Moon - Maia - Kai - Nofish - Bella - Jennifer - Ryan - Katerina - Aiden - Eldric Sage - Mia - Mochi - Bellona - Vincent - Bunny - Neil - Elias - Arthur - Nini - Seren - Pip - Stella - Bodega - Sonrisa - Alek - Dolce - Sohee - Ono Anna - Lenn - Emilien - Andre - Radio Gol - Jada - Dylan - Li - Marcus - Roy - Peter - Sunny - Eric - Rocky - Kiki language: type: string description: >- Synthesis language (maps to Qwen `language_type`). Optional; default `Auto`. **Auto** — use when text mixes languages or the language is uncertain; the model matches pronunciation per segment but accuracy is not guaranteed. **Specific language** — use for single-language text; usually better quality than `Auto`. example: Auto default: Auto enum: - Auto - Chinese - English - German - Italian - Portuguese - Spanish - Japanese - Korean - French - Russian additionalProperties: true securitySchemes: bearerAuth: type: http scheme: bearer description: '`Authorization: Bearer YOUR_API_KEY`' ````