AnyVoiceLab generates speech from a short voice sample and lets users download the resulting audio without logging in.
Qwen3-TTS
Qwen3-TTS generates speech, supports voice design, and includes voice cloning capabilities.
About Qwen3-TTS
Qwen3-TTS is an open-source series of text-to-speech models developed by the Qwen team at Alibaba Cloud. The repository documents released 0.6B and 1.7B models built on the Qwen3-TTS-Tokenizer-12Hz tokenizer, covering speech generation, voice design from free-form descriptions, voice cloning, and natural-language control over tone, speaking rate and emotional expression. A single model handles both streaming and non-streaming generation, with the project citing end-to-end synthesis latency as low as 97 ms.
The models cover ten languages — Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish and Italian — along with multiple dialectal voice profiles. Usage paths listed in the repository include a Python package, a local web UI demo, the DashScope API and vLLM, together with fine-tuning material. Pricing is not stated on the homepage, and no safeguards or consent requirements for voice cloning are described there.
Key features
- Streaming and non-streaming speech generation from one model
- Voice design from free-form user-provided descriptions
- Voice cloning capability
- Instruction control of timbre, emotion and prosody
- Ten languages plus multiple dialectal voice profiles
- Python package, local web UI demo, DashScope API, vLLM usage
Best for: Developers and researchers who need open-source multilingual TTS models for speech generation, voice design and voice cloning.
Good for
- Open-source model series released by the Qwen team at Alibaba Cloud
- Covers ten languages as well as multiple dialectal voice profiles
- One model supports streaming and non-streaming generation with reported latency as low as 97 ms
Watch out for
- No pricing information is stated on the homepage
- Other models mentioned in the technical report are described as not yet released
- The homepage does not describe safeguards or consent requirements for voice cloning
Qwen3-TTS at a glance
- Website
- github.com/QwenLM/Qwen3-TTS
- Category
- AI & Machine Learning › AI Voice Cloning
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
Qwen3-TTS alternatives
Tools and sites you could use instead, ranked by how closely they match.
Fish Audio offers AI text-to-speech and voice cloning with controls for generated emotion.
AIVocal offers voice generation, cloning, editing, speech tools, and transcription for voice related tasks.
VoxWild runs offline on Windows for text-to-speech, with 13 voices, cloning from short recordings and batch generation.
Free
16 alternatives
Uberduck creates AI vocals, text-to-speech, voice conversions, and cloned voices for music and video projects.
Narration Box creates AI voiceovers from text with voice cloning and studio controls across many languages.
Popular in AI Voice Cloning
Well-known listings from the same topic.
Qwen3-TTS FAQ
Which languages does Qwen3-TTS support?
Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish and Italian, plus multiple dialectal voice profiles.
Which Qwen3-TTS models have been released?
The Qwen3-TTS series in 0.6B and 1.7B sizes, based on Qwen3-TTS-Tokenizer-12Hz. Released model entries include Qwen3-TTS-12Hz-1.7B-VoiceDesign and Qwen3-TTS-12Hz-1.7B-CustomVoice.
How can the models be run?
The repository lists a Python package, a local web UI demo, DashScope API usage and vLLM usage, along with fine-tuning material.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.