Loading
Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
MOSS-TTS-Nano (100M, CPU) fine-tunado na voz Iasmin PT-BR. 25.871 clips (11.922 Iasmin real + 13.949 Qwen3-TTS-1.7B voice-clone), 3 epocas. WER heldout 0.049 (Whisper-large-v3). Modo continuation = voz Iasmin nativa...
Model source
Source description
MOSS-TTS-Nano (100M, CPU) fine-tunado na voz Iasmin PT-BR. 25.871 clips (11.922 Iasmin real + 13.949 Qwen3-TTS-1.7B voice-clone), 3 epocas. WER heldout 0.049 (Whisper-large-v3). Modo continuation = voz Iasmin nativa sem audio de referencia.
Sources
1 sourceVerified Sep 21
Model artifacts
1 artifactSource excerpts
2 excerpts--- license: apache-2.0 language: - pt pipeline_tag: text-to-speech base_model: OpenMOSS-Team/MOSS-TTS-Nano tags: - tts - portuguese - iasmin --- # moss-nano-ptbr-iasmin MOSS-TTS-Nano (100M, CPU) fine-tunado na voz Iasmin PT-BR. 25.871 clips (11.922 Iasmin real + 13.949 Qwen3-TTS-1.7B voice-clone), 3 epocas. **WER heldout 0.049** (Whisper-large-v3). Modo continuation = voz Iasmin nativa sem audio de referencia.
Source context: 9 downloads · 0 likes · Pipeline text-to-speech · Repo marcosremar2/moss-nano-ptbr-iasmin
