Support SensevoiceSmall Model

✨ Feature Request

Describe the feature you'd like

Support for the SenseVoiceSmall model (https://huggingface.co/FunAudioLLM/SenseVoiceSmall) as local model.

Problem it solves

A good local Chinese ASR model, and SenseVoiceSmall has been working really well for me in other projects. It'd be great to have it as an option here too since the recognition quality is quite solid.

Proposed solution

Add SenseVoiceSmall as a selectable local model option, similar to how other models are currently supported.

Additional context

• Model: https://huggingface.co/FunAudioLLM/SenseVoiceSmall
• It's optimized for Chinese and works well locally, which is perfect for privacy and offline use