QwenLM / Qwen-Audio

The official repo of Qwen-Audio (通义千问-Audio) chat & pretrained large audio language model proposed by Alibaba Cloud.
Other
1.49k stars 107 forks source link

请问prompt要怎么写才能获得单个task的信息或者想要的task的信息? #32

Closed wjyfelicity closed 8 months ago

wjyfelicity commented 10 months ago

我这边想进行情感识别时,将prompt='<|emotion_recognition|>'时,出来的结果是:

<|emotion_recognition|><|zh|><|transcribe|><|zh|><|notimestamps|><|speaker_meta|>普通话, 女声, 31岁<|startoftranscript|><|zh|><|transcribe|><|zh|><|notimestamps|><|wo_itn|>今天天气真好<|endoftext|> 可以看到上面的结果,包含普通话,性别,年龄和文本,但是就是没有情感,那么写prompt的时候,要怎么写才能获得单个task的信息或者想要的task的信息。
yangjiabupt commented 10 months ago
<|startofanalysis|><|en|><|keyword|><|en|><|notimestamps|><|emotion_recognition|>这样才行
yangjiabupt commented 10 months ago

但是感觉模型的泛化性一般,在新的数据上表现很差的