Downloads · 30 days
32
19% of all-time downloads
robust-speaker-cloning
jatshi/robust-speaker-cloning
robust-speaker-cloning is a text-to-speech model from jatshi. Use it when you need text read aloud. It is set up for cosyvoice. The card lists the license as mit.
Large artifacts for the companion GitHub repository. V2 trains a 6-layer robust speaker Transformer and telephone BWE front end while keeping CosyVoice2-0.5B frozen. On 21 paired real CosyVoice generations, ECAPA x-ve…
All-time downloads
166
Public
Repo size
418 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt213 MB · 51%
At a glance
- Task
- Text-to-Speech
- Library
- cosyvoice
- License
- mit
- Access
- Public
- Created
- Jul 17, 2026
- Updated
- Jul 17, 2026
- SHA
- 9adc9576
Model card
From the Hugging Face model README
Robust Speaker Cloning artifacts
Large artifacts for the companion GitHub repository. V2 trains a 6-layer robust speaker Transformer and telephone BWE front end while keeping CosyVoice2-0.5B frozen. On 21 paired real CosyVoice generations, ECAPA x-vector similarity was 0.6187 for V2 versus 0.5998 baseline (+0.0189 average). This small evaluation also contains negative HVAC and reverb deltas, so it is not a universal-improvement claim.
CosyVoice, CampPlus, ECAPA and AISHELL retain their upstream licenses.
- Task
- Text-to-Speech
- Library
- cosyvoice
- License
- mit
- Languages
- zh
- Created
- Jul 17, 2026
- Updated
- Jul 17, 2026