Sign language ⟶ English, measured.
Upload a continuous, front-facing ASL clip and compare how our models interpret the same video. Real GPU inference. Honest evaluation.
1
Your input & models
Upload ASL clip
{{ vname }}
{{ vmeta }}
Replace
Drag & drop a video file, or click to browse
mp4 · up to 64 MB · 1–40 seconds
Models
(select one or more)
{{ n1 }}
{{ b1T }}
▾
{{ d1 }}
{{ n2 }}
{{ b2T }}
▾
{{ d2 }}
{{ n3 }}
{{ b3T }}
▾
{{ d3 }}
✦{{ trLabel }}
Inference runs on real GPU. No mock results.
2
Results
(one result per model, timed)
Model outputs
Sort by: {{ sortLabel }} ▾
{{ err }}
{{ emptyTitle }}
{{ emptySub }}
{{ r.name }}
{{ r.bT }}
All models ran on the same clip at {{ stamp }}
{{ shareLabel }} ▾
LATEST ROUND{{ roundNum }}{{ roundSub }}
MODELS ONLINE{{ onlineN }} / 3{{ onlineNote }}
AVG. LATENCY (SELECTED){{ avgLat }}Across selected models
Secure processing. Videos are not stored.