{"id":19749,"date":"2025-01-30T15:09:39","date_gmt":"2025-01-30T07:09:39","guid":{"rendered":"https:\/\/www.aisharenet.com\/?p=19749"},"modified":"2025-01-30T15:09:39","modified_gmt":"2025-01-30T07:09:39","slug":"speechgpt-20-preview","status":"publish","type":"post","link":"https:\/\/www.kdjingpai.com\/en\/speechgpt-20-preview\/","title":{"rendered":"SpeechGPT 2.0-preview\uff1a\u5b9e\u65f6\u4ea4\u4e92\u7684\u7aef\u5230\u7aef\u62df\u4eba\u8bed\u97f3\u5bf9\u8bdd\u5927\u6a21\u578b"},"content":{"rendered":"<p>SpeechGPT 2.0-preview \u662f OpenMOSS \u63a8\u51fa\u7684\u9996\u4e2a\u62df\u4eba\u5316\u5b9e\u65f6\u4ea4\u4e92\u7cfb\u7edf\uff0c\u57fa\u4e8e\u767e\u4e07\u5c0f\u65f6\u7ea7\u8bed\u97f3\u6570\u636e\u8bad\u7ec3\u800c\u6210\u3002\u8be5\u7cfb\u7edf\u5177\u5907\u62df\u4eba\u53e3\u8bed\u5316\u8868\u8fbe\u4e0e\u767e\u6beb\u79d2\u7ea7\u4f4e\u5ef6\u8fdf\u54cd\u5e94\uff0c\u652f\u6301\u81ea\u7136\u6d41\u7545\u7684\u5b9e\u65f6\u6253\u65ad\u4ea4\u4e92\u3002SpeechGPT 2.0-preview \u5bf9\u9f50\u4e86\u8bed\u97f3\u548c\u6587\u672c\u4e24\u4e2a\u6a21\u6001\uff0c\u5c55\u73b0\u51fa\u591a\u60c5\u611f\u3001\u591a\u98ce\u683c\u3001\u591a\u97f3\u8272\u7684\u7cbe\u51c6\u63a7\u5236\u4e0e\u667a\u80fd\u5207\u6362\u80fd\u529b\u3002\u5b83\u4e0d\u4ec5\u80fd\u591f\u6a21\u62df\u5404\u7c7b\u89d2\u8272\u7684\u8bed\u6c14\u548c\u60c5\u611f\u72b6\u6001\uff0c\u8fd8\u5177\u5907\u8bd7\u6b4c\u6717\u8bf5\u3001\u6545\u4e8b\u8bb2\u8ff0\u3001\u8bf4\u65b9\u8a00\u7b49\u591a\u79cd\u8bed\u97f3\u624d\u827a\u3002\u6b64\u5916\uff0cSpeechGPT 2.0-preview \u8fd8\u652f\u6301\u5de5\u5177\u8c03\u7528\u3001\u8054\u7f51\u641c\u7d22\u548c\u5916\u6302\u77e5\u8bc6\u5e93\u7b49\u529f\u80fd\uff0c\u63d0\u4f9b\u4e86\u4e30\u5bcc\u7684\u8bed\u97f3\u8868\u73b0\u529b\u548c\u6587\u672c\u80fd\u529b\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter  wp-image-19751\" title=\"SpeechGPT 2.0-preview\uff1a\u5b9e\u65f6\u4ea4\u4e92\u7684\u7aef\u5230\u7aef\u62df\u4eba\u8bed\u97f3\u5bf9\u8bdd\u5927\u6a21\u578b-1\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/0584f0c8dc8f3e5.jpg\" alt=\"SpeechGPT 2.0-preview\uff1a\u5b9e\u65f6\u4ea4\u4e92\u7684\u7aef\u5230\u7aef\u62df\u4eba\u8bed\u97f3\u5bf9\u8bdd\u5927\u6a21\u578b-1\" width=\"1077\" height=\"842\" srcset=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/0584f0c8dc8f3e5.jpg 2552w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/0584f0c8dc8f3e5-300x234.jpg 300w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/0584f0c8dc8f3e5-1024x800.jpg 1024w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/0584f0c8dc8f3e5-768x600.jpg 768w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/0584f0c8dc8f3e5-1536x1200.jpg 1536w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/0584f0c8dc8f3e5-2048x1600.jpg 2048w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/0584f0c8dc8f3e5-15x12.jpg 15w\" sizes=\"auto, (max-width: 1077px) 100vw, 1077px\" \/><\/p>\n<p>&nbsp;<\/p>\n<div id=\"attachment_19750\" style=\"width: 761px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" aria-describedby=\"caption-attachment-19750\" class=\"wp-image-19750 size-full\" title=\"SpeechGPT 2.0-preview\uff1a\u5b9e\u65f6\u4ea4\u4e92\u7684\u7aef\u5230\u7aef\u62df\u4eba\u8bed\u97f3\u5bf9\u8bdd\u5927\u6a21\u578b-1\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/18efa9160ecc9dc.png\" alt=\"SpeechGPT 2.0-preview\uff1a\u5b9e\u65f6\u4ea4\u4e92\u7684\u7aef\u5230\u7aef\u62df\u4eba\u8bed\u97f3\u5bf9\u8bdd\u5927\u6a21\u578b-1\" width=\"751\" height=\"601\" srcset=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/18efa9160ecc9dc.png 751w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/18efa9160ecc9dc-300x240.png 300w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/18efa9160ecc9dc-15x12.png 15w\" sizes=\"auto, (max-width: 751px) 100vw, 751px\" \/><p id=\"caption-attachment-19750\" class=\"wp-caption-text\">\u6f14\u793a\u5730\u5740\uff1ahttps:\/\/sp2.open-moss.com\/<\/p><\/div>\n<p>&nbsp;<\/p>\n<h2>\u529f\u80fd\u5217\u8868<\/h2>\n<ul>\n<li>\u62df\u4eba\u53e3\u8bed\u5316\u8868\u8fbe<\/li>\n<li>\u767e\u6beb\u79d2\u7ea7\u4f4e\u5ef6\u8fdf\u54cd\u5e94<\/li>\n<li>\u591a\u60c5\u611f\u3001\u591a\u98ce\u683c\u3001\u591a\u97f3\u8272\u63a7\u5236<\/li>\n<li>\u89d2\u8272\u626e\u6f14\u80fd\u529b<\/li>\n<li>\u8bd7\u6b4c\u6717\u8bf5\u3001\u6545\u4e8b\u8bb2\u8ff0\u3001\u8bf4\u65b9\u8a00\u7b49\u8bed\u97f3\u624d\u827a<\/li>\n<li>\u652f\u6301\u5de5\u5177\u8c03\u7528\u3001\u8054\u7f51\u641c\u7d22\u548c\u5916\u6302\u77e5\u8bc6\u5e93<\/li>\n<li>\u9ad8\u6548\u8bed\u97f3\u6570\u636e\u722c\u53d6\u7cfb\u7edf<\/li>\n<li>\u591a\u529f\u80fd\u9ad8\u6548\u7387\u8bed\u97f3\u6570\u636e\u6e05\u6d17 pipeline<\/li>\n<li>\u5168\u65b9\u9762\u591a\u7c92\u5ea6\u8bed\u97f3\u6570\u636e\u6807\u6ce8\u7cfb\u7edf<\/li>\n<li>\u8bed\u4e49-\u58f0\u5b66\u8054\u5408\u5efa\u6a21\u7684\u8d85\u4f4e\u6bd4\u7279\u7387\u6d41\u5f0f\u8bed\u97f3 Codec<\/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>\u514b\u9686\u4ed3\u5e93\uff1a<\/li>\n<\/ol>\n<pre><code>   git clone https:\/\/github.com\/OpenMOSS\/SpeechGPT-2.0-preview.git\r\ncd SpeechGPT-2.0-preview\r\n<\/code><\/pre>\n<ol start=\"2\">\n<li>\u4e0b\u8f7d\u6a21\u578b\u6743\u91cd\uff08\u9700\u8981\u5b89\u88c5 git-lfs\uff09\uff1a<\/li>\n<\/ol>\n<pre><code>   git lfs install\r\ngit clone https:\/\/huggingface.co\/fnlp\/SpeechGPT-2.0-preview-Codec\r\ngit clone https:\/\/huggingface.co\/fnlp\/SpeechGPT-2.0-preview-7B\r\n<\/code><\/pre>\n<ol start=\"3\">\n<li>\u51c6\u5907\u73af\u5883\uff1a<\/li>\n<\/ol>\n<pre><code>   pip3 install -r requirements.txt\r\npip3 install flash-attn==2.7.3 --no-build-isolation\r\n<\/code><\/pre>\n<ol start=\"4\">\n<li>\u542f\u52a8\u7f51\u9875 demo\uff1a<\/li>\n<\/ol>\n<pre><code>   python3 demo_gradio.py --codec_ckpt_path SpeechGPT-2.0-preview-Codec\/sg2_codec_ckpt.pkl --model_path SpeechGPT-2.0-preview-7B\/\r\n<\/code><\/pre>\n<h3>\u529f\u80fd\u64cd\u4f5c\u6d41\u7a0b<\/h3>\n<ol>\n<li><strong>\u62df\u4eba\u53e3\u8bed\u5316\u8868\u8fbe<\/strong>\uff1aSpeechGPT 2.0-preview \u80fd\u591f\u6a21\u62df\u4eba\u7c7b\u7684\u53e3\u8bed\u8868\u8fbe\u65b9\u5f0f\uff0c\u63d0\u4f9b\u81ea\u7136\u6d41\u7545\u7684\u5bf9\u8bdd\u4f53\u9a8c\u3002<\/li>\n<li><strong>\u4f4e\u5ef6\u8fdf\u54cd\u5e94<\/strong>\uff1a\u7cfb\u7edf\u5728\u767e\u6beb\u79d2\u7ea7\u522b\u5185\u54cd\u5e94\u7528\u6237\u8f93\u5165\uff0c\u5b9e\u73b0\u5b9e\u65f6\u4ea4\u4e92\u3002<\/li>\n<li><strong>\u591a\u60c5\u611f\u3001\u591a\u98ce\u683c\u3001\u591a\u97f3\u8272\u63a7\u5236<\/strong>\uff1a\u7528\u6237\u53ef\u4ee5\u901a\u8fc7\u6307\u4ee4\u63a7\u5236\u7cfb\u7edf\u7684\u60c5\u611f\u3001\u98ce\u683c\u548c\u97f3\u8272\uff0c\u9002\u5e94\u4e0d\u540c\u7684\u5bf9\u8bdd\u573a\u666f\u3002<\/li>\n<li><strong>\u89d2\u8272\u626e\u6f14<\/strong>\uff1a\u7cfb\u7edf\u80fd\u591f\u6a21\u62df\u4e0d\u540c\u89d2\u8272\u7684\u8bed\u6c14\u548c\u60c5\u611f\u72b6\u6001\uff0c\u9002\u7528\u4e8e\u591a\u79cd\u5e94\u7528\u573a\u666f\u3002<\/li>\n<li><strong>\u8bed\u97f3\u624d\u827a<\/strong>\uff1aSpeechGPT 2.0-preview \u5177\u5907\u8bd7\u6b4c\u6717\u8bf5\u3001\u6545\u4e8b\u8bb2\u8ff0\u548c\u65b9\u8a00\u8868\u8fbe\u7b49\u591a\u79cd\u8bed\u97f3\u624d\u827a\uff0c\u4e30\u5bcc\u4e86\u5bf9\u8bdd\u5185\u5bb9\u3002<\/li>\n<li><strong>\u5de5\u5177\u8c03\u7528\u548c\u8054\u7f51\u641c\u7d22<\/strong>\uff1a\u7cfb\u7edf\u652f\u6301\u8c03\u7528\u5916\u90e8\u5de5\u5177\u548c\u8fdb\u884c\u8054\u7f51\u641c\u7d22\uff0c\u6269\u5c55\u4e86\u5bf9\u8bdd\u7684\u529f\u80fd\u6027\u548c\u4fe1\u606f\u83b7\u53d6\u80fd\u529b\u3002<\/li>\n<li><strong>\u5916\u6302\u77e5\u8bc6\u5e93<\/strong>\uff1a\u901a\u8fc7\u63a5\u5165\u5916\u90e8\u77e5\u8bc6\u5e93\uff0c\u7cfb\u7edf\u80fd\u591f\u63d0\u4f9b\u66f4\u4e3a\u8be6\u5c3d\u548c\u4e13\u4e1a\u7684\u56de\u7b54\u3002<\/li>\n<\/ol>\n<h3>\u4f7f\u7528\u793a\u4f8b<\/h3>\n<ul>\n<li><strong>\u60c5\u611f\u63a7\u5236<\/strong>\uff1a\u7528\u6237\u53ef\u4ee5\u8f93\u5165\u6307\u4ee4\u201c\u7528\u5f00\u5fc3\u7684\u8bed\u6c14\u8bb2\u4e00\u4e2a\u7b11\u8bdd\u201d\uff0c\u7cfb\u7edf\u4f1a\u7528\u6109\u5feb\u7684\u8bed\u6c14\u8bb2\u8ff0\u7b11\u8bdd\u3002<\/li>\n<li><strong>\u89d2\u8272\u626e\u6f14<\/strong>\uff1a\u8f93\u5165\u6307\u4ee4\u201c\u6a21\u62df\u4e00\u4e2a\u8001\u5e08\u7684\u8bed\u6c14\u89e3\u91ca\u4e8c\u6b21\u51fd\u6570\u201d\uff0c\u7cfb\u7edf\u4f1a\u7528\u8001\u5e08\u7684\u8bed\u6c14\u8fdb\u884c\u89e3\u91ca\u3002<\/li>\n<li><strong>\u8bed\u97f3\u624d\u827a<\/strong>\uff1a\u8f93\u5165\u6307\u4ee4\u201c\u7528\u65b9\u8a00\u8bb2\u4e00\u4e2a\u6545\u4e8b\u201d\uff0c\u7cfb\u7edf\u4f1a\u7528\u6307\u5b9a\u7684\u65b9\u8a00\u8bb2\u8ff0\u6545\u4e8b\u3002<\/li>\n<\/ul>\n<p>\u901a\u8fc7\u4ee5\u4e0a\u6b65\u9aa4\u548c\u793a\u4f8b\uff0c\u7528\u6237\u53ef\u4ee5\u5145\u5206\u4f53\u9a8c SpeechGPT 2.0-preview \u7684\u5f3a\u5927\u529f\u80fd\u548c\u591a\u6837\u5316\u5e94\u7528\u573a\u666f\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>SpeechGPT 2.0-preview \u662f OpenMOSS \u63a8\u51fa\u7684\u9996\u4e2a\u62df\u4eba\u5316\u5b9e\u65f6\u4ea4\u4e92\u7cfb\u7edf\uff0c\u57fa\u4e8e\u767e\u4e07\u5c0f\u65f6\u7ea7\u8bed\u97f3\u6570\u636e\u8bad\u7ec3\u800c\u6210\u3002\u8be5\u7cfb\u7edf\u5177\u5907\u62df\u4eba\u53e3\u8bed\u5316\u8868\u8fbe\u4e0e\u767e\u6beb\u79d2\u7ea7\u4f4e\u5ef6\u8fdf\u54cd\u5e94\uff0c\u652f\u6301\u81ea\u7136\u6d41\u7545\u7684\u5b9e\u65f6\u6253\u65ad\u4ea4\u4e92\u3002SpeechGPT 2.0-previ&#8230;<\/p>\n","protected":false},"author":1,"featured_media":61750,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[230,374],"class_list":["post-19749","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-tool","tag-aikaiyuanxiangmu","tag-duandaoduanyuyinzhubei"],"_links":{"self":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/posts\/19749","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/comments?post=19749"}],"version-history":[{"count":0,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/posts\/19749\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/media\/61750"}],"wp:attachment":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/media?parent=19749"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/categories?post=19749"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/tags?post=19749"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}