-
After running the install.sh script, I was only able to download 50 hours of videos out of the 300 hours available. Additionally, the intersection between the downloaded videos and the subtitles is em…
-
I'm currently trying to run the Inference segment on the How2Sign dataset. The pose data given by how2sign is in the form of a json file. The README of GloFE also says to arrange the json files in a c…
-
How to get complete "How2 summarization dataset" data?
-
Hi,
Is dict.all generated from the whole training set of the How2 dataset? But when building a binary dataset for both validation and test set using preprocess.py, there is no token replaced by 'un…
-
## Adding a Dataset
- **Name:** *IWSLT19*
- **Description:** *The Speech Translation Task addresses the translation of English audio into German and Portuguese text.*
- **Hompage:** *https://sites.…
-
When I run run.sh under example2/how，an error occurs:
this url is valid
and I find in #4866 has p…
-
I download the HOW2 dataset,it has many folders,and i can't find files in the conf file.But I can only find files like tran.tok.txt, I can't find files like actions_300.txt ,How should I process the d…
-
# Speech Summarization
Speech summarization refers to the process of condensing spoken language into a shorter version while retaining its essential meaning and key points. Speech summarization aim…
-
# Speech Summary Matching
Speech summarization refers to the process of condensing spoken language into a shorter version while retaining its essential meaning and key points. Speech summarization…
-
I only find several examples of video_action_features. Could you provide the URL of the whole extracted video action feature?