ddlBoJack
Fine-tuning dataset for emotion2vec+ base
Hello, I am curious about the fine-tuning dataset for emotion2vec+ base. Can I assume that the fine-tuning was done using the downstream tasks datasets mentioned in 'Table 1: The datasets at a glance for emotion2vec pre-training and downstream tasks'?
Hi, emotion2vec+ seed is fine-tuned with academic speech emotion data from EmoBox and emotion2vec+ base is fine-tuned with filtered large-scale pseudo-labeled data labeled from emotion2vec+ seed.
Thank you for your response. I would like to know if there are any plans to release the model checkpoints that follow the 'Table 1: The datasets at a glance for emotion2vec pre-training and downstream tasks' setup?
Do you mean the downstream models?你指的是下游模型吗？
Hello! I would like to know which funasr pipeline checkpoints should I use for downstream tasks?
Hello! I was wondering if you can share the training data from EmoBox you used for fine-tuning emotino2vec+ seed. Specifically, how did you split your data into train/val/test?
