search

ai4r / Gesture-Generation-from-Trimodal-Context

Speech Gesture Generation from the Trimodal Context of Text, Audio, and Speaker Identity (SIGGRAPH Asia 2020)
Other
245 stars 35 forks source link

pretrained base line models #14

Closed Sk7w4tch3r closed 3 years ago

Sk7w4tch3r commented 3 years ago

Hi, Can I have the pretrained model for the baseline models? (especially the speech2gesture model)

youngwoo-yoon commented 3 years ago

Hi, I share the pretrained models for seq2seq and speech2gesture. https://kaistackr-my.sharepoint.com/:u:/g/personal/zeroyy_kaist_ac_kr/ESXN1CizG0ZAig9BOkyYkJ8BdXAVKOhNLrSyZW1S494Ffg?e=3tDijB Please be aware that there is the official speech2gesture implementation. https://github.com/amirbar/speech2gesture

Sk7w4tch3r commented 3 years ago

Thanks so much!

Yeah I've seen their work too, but had an issue with their implementation and also wanted to get a sense of the results of the ablation study that you did in your paper. (excluding speech text modality)