gpt-omni / mini-omni

open-source multimodal large language model that can hear, talk while thinking. Featuring real-time end-to-end speech input and streaming audio output conversational capabilities.
https://arxiv.org/abs/2408.16725
MIT License
3.06k stars 273 forks source link

Fix undefined variable 'tik' for non-API mode #27

Closed tengattack closed 2 months ago

tengattack commented 2 months ago

Hi there,

Running without API_URL, it seems omni_gradio.py will load the model at current runtime directly, but it shows following error:

UnboundLocalError: local variable 'tik' referenced before assignment 

This PR fix this.

mini-omni commented 2 months ago

hi, thanks for your contribution.