Download code/static/expressions.html from laion/tyger-speaker-predictor: direct link, hf CLI and curl.
- Browser
- Download file 2.01 kB
-
https://huggingface.co/laion/tyger-speaker-predictor/resolve/main/code/static/expressions.html
- Command line
-
hf download hf://laion/tyger-speaker-predictor/code/static/expressions.html
-
curl -L -o expressions.html https://huggingface.co/laion/tyger-speaker-predictor/resolve/main/code/static/expressions.html
2.01 kB
| <html lang="en"><meta charset="utf-8"><meta name="viewport" content="width=device-width"><title>Robby · 40 expressive poses</title> | |
| <link rel="stylesheet" href="/static/style.css"><link rel="stylesheet" href="/static/expressions.css"> | |
| <header><a href="/">← Talk to Robby</a><div class="brand">40 expressions</div><a href="/speaker-training/">Speaker training ↗</a></header> | |
| <section class="gallery-intro"><h1>One little robot.<br><span>Forty ways to feel.</span></h1><p>Browse the authored EmoNet-Face recipes, blend two expressions and reveal the animation anchors. In conversation, Luna’s TTS acting directions select these poses; this page is a silent preview.</p></section> | |
| <section class="expression-lab"><div id="lab-avatar"></div><div class="lab-controls"><label>Main emotion <select id="a"></select></label><label>Blend with <select id="b"></select></label><label>Blend <output id="blend-value">0%</output><input type="range" id="blend" min="0" max="1" step=".01" value="0"></label><label>Intensity <input type="range" id="intensity" min="0" max="1" step=".01" value="1"></label><label>Natural variation <input type="range" id="variation" min="0" max="1" step=".01" value=".2"></label><button id="seed">New variation</button><label><input id="points" type="checkbox"> Show 101 robot-specific anchors</label><p id="label"></p><p>Continuous brows, eyelids, iris/gaze, mouth and cheeks; eased head, posture and arm gestures. Heart and cross removed, including heart-shaped eye effects.</p></div></section> | |
| <div class="gallery-note">Labels describe intended cartoon expressions, not recognition probabilities. Related social emotions need context; no human 40-way recognition accuracy is claimed. Robot topology is independent of the original French-face rig. <a href="https://github.com/christophschuhmann/cartoon-face-generator">Source recipes ↗</a></div> | |
| <section id="cards" class="expression-grid"></section><script type="module" src="/static/expressions.js"></script></html> | |