metadata
datasets:
- beyoru/vi-voice-woman
language:
- vi
base_model:
- Qwen/Qwen3-Omni-30B-A3B-Instruct
library_name: transformers
pipeline_tag: any-to-any
Overview
- Base Model: Qwen/Qwen3-Omni-30B-A3B-Instruct
- Language: Vietnamese (vi)
- Dataset: beyoru/vi-voice-woman
Description
This model is adapted from Qwen3-Omni to better support Vietnamese speech tasks, with a focus on female vietnamese voice data.
Intended Use
- Vietnamese voice and speech-related applications
- Research and experimentation
Limitations
- Biased toward female voices
- May not generalize well to all Vietnamese accents
- Limitation from dataset so only basic speech can be generate correctly
- Continue update weight (14/01 - 23/01)