xipingL

微调模型，数据中未收录的单词并未影响预测结果

5

Notice: In order to resolve issues more efficiently, please raise issue following the template. （注意：为了更加高效率解决您遇到的问题，请按照模板提问，补充细节） ## ❓ Questions and Help #### What is your question? 你好，我使用paraformer_zh模型进行微调，数据中存在一些不被词典收录的英语单词(类似于funasr)，但是使用微调后的模型预测一些带有这些单词的音频时，并没有得到想要的结果。 想询问如果数据是一些带有专有名词的英语单词，需要怎么微调才能得到想要的结果，能够识别这些专有名词。 #### Code 微调代码没有修改，使用的是https://github.com/modelscope/FunASR/blob/main/examples/industrial_data_pretraining/paraformer/finetune.sh...

起的webui 参数都使用的默认的top_k 15 top_p 1 temperature 1. https://github.com/user-attachments/assets/d06fb59e-fe71-439b-bb1c-dc3a096ea453 prompt_text: Be kind, for everyone you meet is fighting a hard battle. text: Reading is a journey that transcends time and space,...

v2 does not work well with cosyvoice TTS

1

1. use CosyVoice Chinese woman to generate audio (first video), then use OpenVoice ToneColorConverter to generate audio(third video) according target_se(second video) that has serious electrical tone 2. use CosyVoice Chinese...