Alibaba · audio
Qwen3 TTS 1.7B CustomVoice
Qwen3 TTS 1.7B CustomVoice is a downloadable text-to-speech checkpoint with preset speakers and instruction-based voice control. It is distinct from the Base voice-cloning and VoiceDesign checkpoints.
Downloadable checkpoint · apache-2.0
Official sources reviewed 2026-10-02. This page documents access and specifications; it does not report a model benchmark.
Task and deployment contract
Downloadable checkpoint; run the task-specific inference code linked below.
Use generate_custom_voice with the documented language, speaker and optional instruction arguments. Query supported speakers instead of inventing voice IDs.
The published checkpoint supports Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish and Italian.
CustomVoice provides preset speakers. Voice cloning belongs to the Base checkpoint and voice creation from descriptions belongs to VoiceDesign; these are separate assets and APIs.
The checkpoint card lists Apache-2.0. Hosting hardware, serving code, input voice permissions and output review remain separate operational requirements.
Billing and limits
No current provider rate for this exact configuration is recorded here. Open weights still require compute and operations; an unlisted rate does not mean a hosted API is free.
Audio limits are task-specific; no text-token context window is inferred.
Fast Pivot — Access hundreds of AI models from a single API — competitive pricing, reliable uptime, enterprise-grade infrastructure.