Expired50% confidenceFactExact time
CosyVoice v3.5 supports a voice design feature that allows users to generate entirely new timbres through natural language descriptions without providing reference audio.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Related Claims
Partially VerifiedCosyVoice is an open-source speech synthesis model from Alibaba's Tongyi Lab that supports zero-shot voice cloning and can generate a target voice timbre using just a few seconds of reference audio.78% similarUnverifiedCosyVoice v3.5 features a HotFix pronunciation correction function that can directly specify the reading of polyphonic characters or replace special pronunciations.78% similarUnverifiedCosyVoice v3.5 Plus supports free-form voice instruction control but does not support system-preset voice profiles, requiring users to design voices themselves77% similarUnverifiedCoze supports voice reply (text-to-speech with selectable voice) and cloud calling (real-time voice conversation) features.72% similarUnverified该项目支持声纹识别和声音克隆功能68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/46244API
curl https://kongchang.com/api/v1/knowledge/claims/46244MCP
get_claim(id=46244)