American-Sign-Language-Dataset
Verified108,618 videos of 2,208 ASL words for gesture recognition.
What is American-Sign-Language-Dataset?
The dataset provides a large collection of preprocessed ASL videos suitable for video-based machine learning.
It is useful for researchers and developers building models for sign language recognition and gesture detection tasks.
What you can build with American-Sign-Language-Dataset
Sign Language Recognition Model Training
Train computer vision models to classify ASL words from video clips for real-time gesture detection systems.
ASL-to-Text Translation Pipeline
Build sequence-to-sequence models that convert sign language video input into English text output.
Interactive ASL Learning App
Develop feedback tools that compare user-recorded signs against dataset examples to score accuracy.
Load American-Sign-Language-Dataset
from datasets import load_dataset
ds = load_dataset("just-me7ss/American-Sign-Language-Dataset")- 1pip install datasets
- 2from datasets import load_dataset
- 3ds = load_dataset('just-me7ss/American-Sign-Language-Dataset')
- 4Inspect splits, video paths, and word labels
- 5Download and preprocess videos locally for training
American-Sign-Language-Dataset: pros & cons
Pros
- +Over 108k videos across 2208 words
- +Minimum 30 examples per word
- +Preprocessed for consistency
- +Direct Hugging Face access
Cons
- –Video storage can exceed several GB
- –Category listed as nlp-text but contains video
- –Quality may vary across source videos
Frequently asked questions
A Hugging Face dataset of 108,618 preprocessed ASL videos covering 2,208 words with at least 30 videos each.
User reviews
Verified reviews from the community shape this listing's rating.
Loading reviews…
Similar datasets
Other text & nlp options worth comparing.