Unverified50% confidenceFactExact time
The uploader mainly uses VoxCPM2 as the voice synthesis model, while also retaining Qwen3-TTS as another comparable route.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/1/2026
First Seen
Valid until: 1/30/2027
Sources
Related Entities
Related Claims
UnverifiedQwen-Audio-3.0-TTS can output clean synthesis results even when the reference audio contains noise.74% similarUnverifiedUsers can load audio models such as Voxtral and Qwen2-Audio via FastModel.73% similarUnverifiedNari has launched two speech models based on Qwen3: a text-to-speech (TTS) model and an automatic speech recognition (ASR) model70% similarUnverifiedThe voice stack consists of four core components: Qwen 3.6 35B (a mixture-of-experts model), Whisper, Breeze TTS2 (3.5B parameters), and Celery VAD.70% similarUnverifiedThe uploader's song-request feature uses VoxCPM2 to perform timbre conversion on existing vocal performances, saving the processed clips to a local playlist, and plays the processed version when a song is requested rather than generating a new song from scratch on the spot.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1124832API
curl https://kongchang.com/api/v1/knowledge/claims/1124832MCP
get_claim(id=1124832)