Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.platform.qubrid.com/llms.txt

Use this file to discover all available pages before exploring further.

​
About the Provider

Alibaba Cloud is the cloud computing arm of Alibaba Group and the creator of the Qwen model family. Through its open-source initiative, Alibaba has released state-of-the-art language and multimodal models under permissive licenses, enabling developers and enterprises to build powerful AI applications across diverse domains and languages.

​
Model Quickstart

This section helps you quickly get started with the qwen3-tts-flash model on the Qubrid AI inferencing platform. To use this model, you need:
  • A valid Qubrid API key
  • Access to the Qubrid inference API
  • Basic knowledge of making API requests in your preferred language
Once authenticated with your API key, you can send inference requests to the qwen3-tts-flash model and receive responses based on your input prompts. Below are example placeholders showing how the model can be accessed using different programming environments.
You can choose the one that best fits your workflow. 
import requests
import json

url = "https://platform.qubrid.com/v1/audio/generations"
headers = {
    "Authorization": "Bearer QUBRID_API_KEY",
    "Content-Type": "application/json",
}

data = {
    "model": "qwen3-tts-flash",
    "text": "Today is a wonderful day to build something people love!",
    "voice": "Cherry",
    "language_type": "Auto",
}

response = requests.post(url, headers=headers, json=data)
print(response.json())
Always match your text language with the language_type you set. Sending English text with language_type: Chinese will produce broken or unnatural audio. The model speaks best when the script and language are aligned.

​
Available Voices

“Welcome to Qubrid. This demo shows how easy it is to turn text into natural speech.”
Explore the voices — find the one that brings your product to life.
Voice selection matters. A few confirmed highlights:
  • Arthur — Bold, deep, authoritative male voice. Best for trailers, audiobooks, documentary narration, and cinematic scripts.
  • Nini — Expressive, anime-style voice. Best for game NPCs, visual novel characters, and animated dubs.
  • Stella — Anime-style voice with a softer tone. Best for companion characters and emotional scenes.
Experiment with all available voices on the Qubrid Playground

​
Voice Scripts & Use Cases

Qwen3-TTS-Flash isn’t just a text reader — it’s a production-grade voice engine. Below are real scripts across languages and use cases that show exactly what this model can do. Copy any of these directly into your API call.

​
🇺🇸 English — Customer Support Agent

Use case: Automated voice response for a SaaS support chatbot 
"Thank you for reaching out to Qubrid Support. I'm looking into your account right now.
It looks like your API usage spiked earlier today — don't worry, your limits have been reset
and your service is fully active. Is there anything else I can help you with today?"

{
  "model": "qwen3-tts-flash",
  "text": "Thank you for reaching out to Qubrid Support. I'm looking into your account right now. It looks like your API usage spiked earlier today — don't worry, your limits have been reset and your service is fully active. Is there anything else I can help you with today?",
  "voice": "Cherry",
  "language_type": "English"
}

​
🇨🇳 Chinese — E-Commerce Onboarding Voiceover

Use case: Product walkthrough narration for a Chinese marketplace app 
"欢迎来到 Qubrid AI 平台！在这里，您可以轻松访问全球顶尖的 AI 模型，
无需任何基础设施搭建。只需一个 API 密钥，即可立刻开始体验。
让我们一起开启智能应用的新时代！"

{
  "model": "qwen3-tts-flash",
  "text": "欢迎来到 Qubrid AI 平台！在这里，您可以轻松访问全球顶尖的 AI 模型，无需任何基础设施搭建。只需一个 API 密钥，即可立刻开始体验。让我们一起开启智能应用的新时代！",
  "voice": "Cherry",
  "language_type": "Chinese"
}

​
🇧🇷 Portuguese — Interactive Chatbot Response

Use case: Voice-enabled virtual assistant for a Brazilian fintech app 
"Olá! Seja bem-vindo ao seu assistente financeiro pessoal.
Seu saldo atual é de R$ 4.250,00 e você tem duas faturas vencendo esta semana.
Gostaria que eu agendasse os pagamentos automaticamente para você?"

{
  "model": "qwen3-tts-flash",
  "text": "Olá! Seja bem-vindo ao seu assistente financeiro pessoal. Seu saldo atual é de R$ 4.250,00 e você tem duas faturas vencendo esta semana. Gostaria que eu agendasse os pagamentos automaticamente para você?",
  "voice": "Cherry",
  "language_type": "Portuguese"
}

​
🎌 Anime & Gaming — Character Voice Script

Use case: Dynamic NPC dialogue generation for a Japanese-style game or anime dub 
"W-wait, you actually came back for me? I thought... I thought you forgot.
Nobody ever comes back. But you did. Maybe... maybe I can trust you after all.
Just don't make me regret this, okay?"

{
  "model": "qwen3-tts-flash",
  "text": "W-wait, you actually came back for me? I thought... I thought you forgot. Nobody ever comes back. But you did. Maybe... maybe I can trust you after all. Just don't make me regret this, okay?",
  "voice": "Nini",
  "language_type": "English"
}
💡 Pro tip: Nini and Stella carry that expressive, emotional anime-style tone — perfect for game characters, visual novel dialogue, and animated dubs. Try Stella for a softer, gentler character and Nini for something more spirited and reactive.

​
Model Overview

Qwen3-TTS-Flash is a fast, high-quality text-to-speech model supporting multiple voices, languages, and expressive speaking styles.
  • It is ideal for real-time applications and interactive experiences, built on a neural TTS architecture with transformer-based acoustic modeling and vocoder.
  • With multilingual support, configurable voice and language options, and low-latency synthesis, it is suitable for a wide range of production audio generation workflows.

​
Model at a Glance

FeatureDetails
Model IDqwen3-tts-flash
ProviderAlibaba Cloud (Qwen Team)
ArchitectureNeural TTS with transformer-based acoustic modeling and vocoder
Model SizeMulti-billion parameters (approx.)
Context LengthUp to ~10K characters
Release Date2025
LicenseApache 2.0
Training DataN/A

​
When to use?

You should consider using Qwen3 TTS Flash if:
  • You need product tutorials and onboarding voiceovers generated from documentation or scripts
  • Your application requires voice-enabling chatbots and virtual assistants for a more engaging UX
  • You are generating narration for marketing videos, explainers, and social content
  • Your use case involves accessibility features such as screen-reading and audio summaries of long text
  • You need educational content, audiobooks, and podcast-like experiences generated from text

​
Supported Languages

Languagelanguage_type
Auto-detectAuto
ChineseChinese
EnglishEnglish
GermanGerman
ItalianItalian
PortuguesePortuguese
SpanishSpanish
JapaneseJapanese
KoreanKorean
FrenchFrench
RussianRussian

​
Inference Parameters

Parameter NameTypeDefaultDescription
VoiceselectCherrySelect the speaker voice for synthesis.
LanguageselectAutoLanguage hint for the TTS request.

​
Key Features

  • Multiple High-Quality Voices: A selection of expressive, natural-sounding speaker voices for diverse use cases.
  • Multilingual Support: Handles multiple languages with automatic language detection.
  • Low Latency Synthesis: Optimized for real-time audio generation and interactive applications.
  • Configurable Voice and Language: Flexible voice and language selection per request.
  • Apache 2.0 License: Fully open-source with unrestricted commercial use.

​
Summary

Qwen3-TTS-Flash is Alibaba’s fast text-to-speech model built for real-time multilingual audio synthesis.
  • It uses a neural TTS architecture with transformer-based acoustic modeling and vocoder, supporting multiple expressive voices.
  • It is optimized for low-latency synthesis across product voiceovers, chatbot audio, accessibility features, and educational content.
  • The model supports configurable voice and language settings with up to ~10K character context.
  • Licensed under Apache 2.0 for full commercial use.