{"id":13086,"date":"2024-11-07T17:44:07","date_gmt":"2024-11-07T09:44:07","guid":{"rendered":"https:\/\/www.aisharenet.com\/?p=13086"},"modified":"2025-01-11T12:22:56","modified_gmt":"2025-01-11T04:22:56","slug":"ultralight-digital-human","status":"publish","type":"post","link":"https:\/\/www.kdjingpai.com\/ja\/ultralight-digital-human\/","title":{"rendered":"UltraLight Digital Human\uff1a\u5f00\u6e90\u7aef\u4fa7\u5b9e\u65f6\u8fd0\u884c\u7684\u8d85\u8f7b\u91cf\u7ea7\u6570\u5b57\u4eba\uff0c\u9644\u4e00\u952e\u5b89\u88c5\u5305"},"content":{"rendered":"<p>Ultralight Digital Human \u662f\u4e00\u4e2a\u5f00\u6e90\u9879\u76ee\uff0c\u65e8\u5728\u5f00\u53d1\u4e00\u4e2a\u8d85\u8f7b\u91cf\u7ea7\u7684\u6570\u5b57\u4eba\u6a21\u578b\uff0c\u53ef\u4ee5\u5728\u79fb\u52a8\u8bbe\u5907\u4e0a\u5b9e\u65f6\u8fd0\u884c\u3002\u8be5\u9879\u76ee\u901a\u8fc7\u4f18\u5316\u7b97\u6cd5\u548c\u6a21\u578b\u7ed3\u6784\uff0c\u5b9e\u73b0\u4e86\u5728\u79fb\u52a8\u7aef\u8bbe\u5907\u4e0a\u7684\u6d41\u7545\u8fd0\u884c\uff0c\u9002\u7528\u4e8e\u793e\u4ea4\u5e94\u7528\u3001\u6e38\u620f\u548c\u865a\u62df\u73b0\u5b9e\u7b49\u591a\u79cd\u573a\u666f\u3002\u7528\u6237\u53ef\u4ee5\u8f7b\u677e\u8bad\u7ec3\u548c\u90e8\u7f72\u81ea\u5df1\u7684\u6570\u5b57\u4eba\u6a21\u578b\uff0c\u4eab\u53d7\u4e2a\u6027\u5316\u548c\u6c89\u6d78\u5f0f\u7684\u4f53\u9a8c\u3002<\/p>\n<p>\u5173\u4e8e\u5728\u79fb\u52a8\u7aef\u4e0a\u8fd0\u884c\u4e5f\u662f\u6ca1\u95ee\u9898\u7684\uff0c\u53ea\u9700\u8981\u628a\u73b0\u5728\u8fd9\u4e2a\u6a21\u578b\u901a\u9053\u6570\u6539\u5c0f\u4e00\u70b9\uff0c\u97f3\u9891\u7279\u5f81\u7528wenet\u5c31\u6ca1\u95ee\u9898\u4e86\u3002<\/p>\n<p><img decoding=\"async\" title=\"UltraLight Digital Human\uff1a\u624b\u673a\u8bbe\u5907\u5373\u53ef\u5b9e\u65f6\u8fd0\u884c\u7684\u8d85\u8f7b\u91cf\u7ea7\u6570\u5b57\u4eba\u6a21\u578b-1\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/2124b35560f77d3.png\" alt=\"UltraLight Digital Human\uff1a\u624b\u673a\u8bbe\u5907\u5373\u53ef\u5b9e\u65f6\u8fd0\u884c\u7684\u8d85\u8f7b\u91cf\u7ea7\u6570\u5b57\u4eba\u6a21\u578b-1\" \/><\/p>\n<p>&nbsp;<\/p>\n<h2>\u529f\u80fd\u5217\u8868<\/h2>\n<ul>\n<li><strong>\u5b9e\u65f6\u8fd0\u884c<\/strong>\uff1a\u6a21\u578b\u53ef\u4ee5\u5728\u79fb\u52a8\u8bbe\u5907\u4e0a\u5b9e\u65f6\u8fd0\u884c\uff0c\u54cd\u5e94\u8fc5\u901f\u3002<\/li>\n<li><strong>\u8f7b\u91cf\u7ea7\u8bbe\u8ba1<\/strong>\uff1a\u4f18\u5316\u540e\u7684\u6a21\u578b\u7ed3\u6784\uff0c\u9002\u5408\u8d44\u6e90\u6709\u9650\u7684\u79fb\u52a8\u8bbe\u5907\u3002<\/li>\n<li><strong>\u5f00\u6e90\u9879\u76ee<\/strong>\uff1a\u4ee3\u7801\u548c\u6a21\u578b\u5b8c\u5168\u5f00\u6e90\uff0c\u7528\u6237\u53ef\u4ee5\u81ea\u7531\u4fee\u6539\u548c\u4f7f\u7528\u3002<\/li>\n<li><strong>\u591a\u573a\u666f\u5e94\u7528<\/strong>\uff1a\u9002\u7528\u4e8e\u793e\u4ea4\u5e94\u7528\u3001\u6e38\u620f\u3001\u865a\u62df\u73b0\u5b9e\u7b49\u591a\u79cd\u573a\u666f\u3002<\/li>\n<li><strong>\u97f3\u9891\u7279\u5f81\u63d0\u53d6<\/strong>\uff1a\u652f\u6301 wenet \u548c hubert \u4e24\u79cd\u97f3\u9891\u7279\u5f81\u63d0\u53d6\u65b9\u6848\u3002<\/li>\n<li><strong>\u540c\u6b65\u7f51\u7edc<\/strong>\uff1a\u901a\u8fc7 syncnet \u6280\u672f\u63d0\u5347\u5507\u5f62\u540c\u6b65\u6548\u679c\u3002<\/li>\n<li><strong>\u8be6\u7ec6\u6559\u7a0b<\/strong>\uff1a\u63d0\u4f9b\u8be6\u7ec6\u7684\u8bad\u7ec3\u548c\u4f7f\u7528\u6559\u7a0b\uff0c\u5e2e\u52a9\u7528\u6237\u5feb\u901f\u4e0a\u624b\u3002<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h2>\u4f7f\u7528\u5e2e\u52a9<\/h2>\n<h3>\u5b89\u88c5\u6d41\u7a0b<\/h3>\n<ol>\n<li><strong>\u73af\u5883\u51c6\u5907<\/strong>\uff1a\n<ul>\n<li>\u5b89\u88c5 Python 3.10 \u53ca\u4ee5\u4e0a\u7248\u672c\u3002<\/li>\n<li>\u5b89\u88c5 PyTorch 1.13.1 \u53ca\u5176\u4ed6\u4f9d\u8d56\u5e93\uff1a\n<pre><code>conda create -n dh python=3.10\r\nconda activate dh\r\nconda install pytorch==1.13.1 torchvision==0.14.1 torchaudio==0.13.1 pytorch-cuda=11.7 -c pytorch -c nvidia\r\nconda install mkl=2024.0\r\npip install opencv-python transformers numpy==1.23.5 soundfile librosa onnxruntime\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u4e0b\u8f7d\u6a21\u578b\u6587\u4ef6<\/strong>\uff1a\n<ul>\n<li>\u4ece\u4ee5\u4e0b\u94fe\u63a5\u4e0b\u8f7d wenet encoder.onnx \u6587\u4ef6\uff0c\u5e76\u653e\u7f6e\u5728 <code>data_utils\/<\/code> \u76ee\u5f55\u4e0b\uff1a <a href=\"https:\/\/drive.google.com\/file\/d\/1e4Z9zS053JEWl6Mj3W9Lbc9GDtzHIg6b\/view?usp=drive_link\" target=\"_blank\" rel=\"noopener\">\u4e0b\u8f7d\u94fe\u63a5<\/a><\/li>\n<\/ul>\n<\/li>\n<\/ol>\n<h3>\u4f7f\u7528\u6d41\u7a0b<\/h3>\n<ol>\n<li><strong>\u51c6\u5907\u89c6\u9891<\/strong>\uff1a\n<ul>\n<li>\u51c6\u5907\u4e00\u4e2a 3-5 \u5206\u949f\u7684\u89c6\u9891\uff0c\u786e\u4fdd\u6bcf\u4e00\u5e27\u90fd\u6709\u5b8c\u6574\u7684\u9762\u90e8\u66dd\u5149\uff0c\u58f0\u97f3\u6e05\u6670\u65e0\u6742\u97f3\u3002<\/li>\n<li>\u5c06\u89c6\u9891\u653e\u7f6e\u5728\u4e00\u4e2a\u65b0\u7684\u6587\u4ef6\u5939\u4e2d\u3002<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u63d0\u53d6\u97f3\u9891\u7279\u5f81<\/strong>\uff1a\n<ul>\n<li>\u4f7f\u7528\u4ee5\u4e0b\u547d\u4ee4\u63d0\u53d6\u97f3\u9891\u7279\u5f81\uff1a\n<pre><code>cd data_utils\r\npython process.py YOUR_VIDEO_PATH --asr hubert\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u8bad\u7ec3\u6a21\u578b<\/strong>\uff1a\n<ul>\n<li>\u8bad\u7ec3 syncnet \u6a21\u578b\u4ee5\u83b7\u5f97\u66f4\u597d\u7684\u6548\u679c\uff1a\n<pre><code>cd ..\r\npython syncnet.py --save_dir .\/syncnet_ckpt\/ --dataset_dir .\/data_dir\/ --asr hubert\r\n<\/code><\/pre>\n<\/li>\n<li>\u4f7f\u7528\u6700\u4f4e loss \u7684 checkpoint \u8bad\u7ec3\u6570\u5b57\u4eba\u6a21\u578b\uff1a\n<pre><code>python train.py --dataset_dir .\/data_dir\/ --save_dir .\/checkpoint\/ --asr hubert --use_syncnet --syncnet_checkpoint syncnet_ckpt\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u63a8\u7406<\/strong>\uff1a\n<ul>\n<li>\u63d0\u53d6\u6d4b\u8bd5\u97f3\u9891\u7279\u5f81\uff1a\n<pre><code>python extract_test_audio.py YOUR_TEST_AUDIO_PATH --asr hubert\r\n<\/code><\/pre>\n<\/li>\n<li>\u8fd0\u884c\u63a8\u7406\uff1a\n<pre><code>python inference.py --dataset .\/data_dir\/ --audio_feat .\/your_test_audio_hu.npy --save_path .\/output.mp4 --checkpoint .\/checkpoint\/best_model.pth\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n<h3>\u6ce8\u610f\u4e8b\u9879<\/h3>\n<ul>\n<li>\u786e\u4fdd\u89c6\u9891\u5e27\u7387\u4e0e\u6240\u9009\u7684\u97f3\u9891\u7279\u5f81\u63d0\u53d6\u65b9\u6848\u5339\u914d\uff1awenet \u4e3a 20fps\uff0chubert \u4e3a 25fps\u3002<\/li>\n<li>\u5728\u8bad\u7ec3\u548c\u63a8\u7406\u8fc7\u7a0b\u4e2d\uff0c\u5b9a\u671f\u76d1\u63a7\u6a21\u578b\u7684 loss \u503c\uff0c\u9009\u62e9\u6700\u4f18\u7684 checkpoint \u8fdb\u884c\u8bad\u7ec3\u3002<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h2>\u7334\u54e5\u4e00\u952e\u6574\u5408\u5305<\/h2>\n<p><div class=\"huoduan_hide_box\" style=\"border:1px dashed #F60; padding:10px; margin:10px 0; line-height:200%; color:#F00; background-color:#FFF4FF; overflow:hidden; clear:both;\"><img loading=\"lazy\" decoding=\"async\" class=\"wxpic\" align=\"right\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/05\/d8668ed8023fbe2.jpg\" style=\"width:150px;height:150px;margin-left:20px;display:inline;border:none\" width=\"150\" height=\"150\"  alt=\"AI\u751f\u4ea7\u529b\u5e94\u7528\" \/><span style=\"font-size:18px;\">\u6b64\u5904\u5185\u5bb9\u5df2\u7ecf\u88ab\u4f5c\u8005\u9690\u85cf\uff0c\u8bf7\u8f93\u5165\u9a8c\u8bc1\u7801\u67e5\u770b\u5185\u5bb9<\/span><form method=\"post\" style=\"margin:10px 0;\"><span class=\"yzts\" style=\"font-size:18px;float:left;\">\u9a8c\u8bc1\u7801\uff1a<\/span><input name=\"huoduan_verifycode\" id=\"verifycode\" type=\"text\" value=\"\" style=\"border:none;float:left;width:80px; height:32px; line-height:30px; padding:0 5px; border:1px solid #FF6600;-moz-border-radius: 0px;  -webkit-border-radius: 0px;  border-radius:0px;\" \/><input id=\"verifybtn\" style=\"border:none;float:left;width:80px; height:32px; line-height:32px; padding:0 5px; background-color:#F60; text-align:center; border:none; cursor:pointer; color:#FFF;-moz-border-radius: 0px; font-size:14px;  -webkit-border-radius: 0px;  border-radius:0px;\" name=\"\" type=\"submit\" value=\"\u63d0\u4ea4\u67e5\u770b\" \/><\/form><div style=\"clear:left;\"><\/div><span style=\"color:#00BF30\">\u8bf7\u5173\u6ce8\u672c\u7ad9\u5fae\u4fe1\u516c\u4f17\u53f7\uff0c\u56de\u590d\u201c<span style=\"color:blue\">\u9a8c\u8bc1\u7801<\/span>\u201d\uff0c\u83b7\u53d6\u9a8c\u8bc1\u7801\u3002\u5728\u5fae\u4fe1\u91cc\u641c\u7d22\u201c<span style=\"color:blue\">AI\u751f\u4ea7\u529b\u5e94\u7528<\/span>\u201d\u6216\u8005\u201c<span style=\"color:blue\">Artificial9527<\/span>\u201d\u6216\u8005\u5fae\u4fe1\u626b\u63cf\u53f3\u4fa7\u4e8c\u7ef4\u7801\u90fd\u53ef\u4ee5\u5173\u6ce8\u672c\u7ad9\u5fae\u4fe1\u516c\u4f17\u53f7\u3002<\/span><div class=\"cl\"><\/div><\/div><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Ultralight Digital Human \u662f\u4e00\u4e2a\u5f00\u6e90\u9879\u76ee\uff0c\u65e8\u5728\u5f00\u53d1\u4e00\u4e2a\u8d85\u8f7b\u91cf\u7ea7\u7684\u6570\u5b57\u4eba\u6a21\u578b\uff0c\u53ef\u4ee5\u5728\u79fb\u52a8\u8bbe\u5907\u4e0a\u5b9e\u65f6\u8fd0\u884c\u3002\u8be5\u9879\u76ee\u901a\u8fc7\u4f18\u5316\u7b97\u6cd5\u548c\u6a21\u578b\u7ed3\u6784\uff0c\u5b9e\u73b0\u4e86\u5728\u79fb\u52a8\u7aef\u8bbe\u5907\u4e0a\u7684\u6d41\u7545\u8fd0\u884c\uff0c\u9002\u7528\u4e8e\u793e\u4ea4\u5e94\u7528\u3001\u6e38\u620f\u548c\u865a\u62df\u73b0\u5b9e\u7b49\u591a\u79cd\u573a\u666f\u3002\u7528\u6237\u53ef\u4ee5\u8f7b&#8230;<\/p>\n","protected":false},"author":1,"featured_media":61207,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[228],"class_list":["post-13086","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-tool","tag-aishuziren"],"_links":{"self":[{"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/posts\/13086","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/comments?post=13086"}],"version-history":[{"count":0,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/posts\/13086\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/media\/61207"}],"wp:attachment":[{"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/media?parent=13086"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/categories?post=13086"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/tags?post=13086"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}