vibevoice-asr-it-yodas-only

Fine-tuned microsoft/VibeVoice-ASR on YODAS-Granary Italian (asr_only ~87k + ast 200k cap, conversational YouTube audio).

The LoRA adapter has been merged into the base model — drop-in replacement.

Usage

from vibevoice.modular.modeling_vibevoice_asr import VibeVoiceASRForConditionalGeneration
from vibevoice.processor.vibevoice_asr_processor import VibeVoiceASRProcessor

processor = VibeVoiceASRProcessor.from_pretrained("LocalAI-io/vibevoice-asr-it-yodas-only")
model = VibeVoiceASRForConditionalGeneration.from_pretrained("LocalAI-io/vibevoice-asr-it-yodas-only", dtype="bfloat16")

For just the LoRA adapter, see LocalAI-io/vibevoice-asr-it-yodas-only-lora.