FunAudioLLM / SenseVoice

Multilingual Voice Understanding Model
https://funaudiollm.github.io/
Other
2.67k stars 254 forks source link

请问情绪识别功能怎么使用,有相应的推理脚本吗 #11

Closed HeJinLing closed 2 months ago

HeJinLing commented 2 months ago

加载模型参数文件,会报错需要tokenizer的token_list(参数文件中没有,且没有对应的输入格式) 另外SER入参应该在何处送入呢

joseph16388 commented 2 months ago

顶!

LauraGPT commented 2 months ago

What's your problem?

gaochangfeng commented 2 months ago

默认推理参数下模型输出的前三个token分别为语种、情感和事件,不需要专门配置