{"id":13082,"date":"2024-11-07T16:35:14","date_gmt":"2024-11-07T08:35:14","guid":{"rendered":"https:\/\/www.aisharenet.com\/?p=13082"},"modified":"2025-07-19T05:37:49","modified_gmt":"2025-07-18T21:37:49","slug":"outetts","status":"publish","type":"post","link":"https:\/\/www.kdjingpai.com\/ja\/outetts\/","title":{"rendered":"OuteTTS\uff1a\u5b9e\u9a8c\u6027\u6587\u672c\u8f6c\u8bed\u97f3\u6a21\u578b\uff0c\u4f7f\u7528\u7eaf\u8bed\u8a00\u5efa\u6a21\u65b9\u6cd5\u5b9e\u73b0\u7684TTS"},"content":{"rendered":"<p>OuteTTS \u662f\u4e00\u79cd\u5b9e\u9a8c\u6027\u6587\u672c\u8f6c\u8bed\u97f3\uff08TTS\uff09\u6a21\u578b\uff0c\u91c7\u7528\u7eaf\u8bed\u8a00\u5efa\u6a21\u65b9\u6cd5\u751f\u6210\u9ad8\u8d28\u91cf\u8bed\u97f3\u3002\u4e0e\u4f20\u7edf\u7684 TTS \u7cfb\u7edf\u4e0d\u540c\uff0cOuteTTS \u4e0d\u9700\u8981\u5916\u90e8\u9002\u914d\u5668\u6216\u590d\u6742\u7684\u67b6\u6784\u3002\u8be5\u6a21\u578b\u57fa\u4e8e LLaMa \u67b6\u6784\uff0c\u652f\u6301\u8bed\u97f3\u514b\u9686\u529f\u80fd\uff0c\u80fd\u591f\u751f\u6210\u5177\u6709\u968f\u673a\u8bf4\u8bdd\u4eba\u7279\u5f81\u7684\u8bed\u97f3\u3002OuteTTS \u65e8\u5728\u901a\u8fc7\u7b80\u5355\u7684\u7ed3\u6784\u5b9e\u73b0\u9ad8\u6548\u7684\u8bed\u97f3\u5408\u6210\uff0c\u9002\u7528\u4e8e\u591a\u79cd\u5e94\u7528\u573a\u666f\u3002<\/p>\n<blockquote><p>OuteTTS-0.1-350M \u5728\u7b80\u5316\u6587\u672c\u5230\u8bed\u97f3\u5408\u6210\u65b9\u9762\u5411\u524d\u8fc8\u8fdb\u4e86\u4e00\u6b65\u3002 OuteTTS-0.1-350M \u8bc1\u660e\u4e86\u901a\u8fc7\u7eaf\u7cb9\u7684\u8bed\u8a00\u5efa\u6a21\u65b9\u6cd5\u53ef\u4ee5\u751f\u6210\u9ad8\u8d28\u91cf\u7684\u8bed\u97f3\u3002<\/p><\/blockquote>\n<p>&nbsp;<\/p>\n<h2>\u529f\u80fd\u5217\u8868<\/h2>\n<ul>\n<li><strong>\u6587\u672c\u8f6c\u8bed\u97f3<\/strong>\uff1a\u5c06\u8f93\u5165\u7684\u6587\u672c\u8f6c\u6362\u4e3a\u81ea\u7136\u6d41\u7545\u7684\u8bed\u97f3\u3002<\/li>\n<li><strong>\u8bed\u97f3\u514b\u9686<\/strong>\uff1a\u901a\u8fc7\u53c2\u8003\u97f3\u9891\u6587\u4ef6\u521b\u5efa\u81ea\u5b9a\u4e49\u8bf4\u8bdd\u4eba\uff0c\u5e76\u751f\u6210\u76f8\u5e94\u7684\u8bed\u97f3\u3002<\/li>\n<li><strong>\u591a\u6a21\u578b\u652f\u6301<\/strong>\uff1a\u652f\u6301 Hugging Face \u6a21\u578b\u548c GGUF \u6a21\u578b\u3002<\/li>\n<li><strong>\u97f3\u9891\u64ad\u653e\u4e0e\u4fdd\u5b58<\/strong>\uff1a\u751f\u6210\u7684\u8bed\u97f3\u53ef\u4ee5\u76f4\u63a5\u64ad\u653e\u6216\u4fdd\u5b58\u4e3a\u97f3\u9891\u6587\u4ef6\u3002<\/li>\n<li><strong>\u6e29\u5ea6\u4e0e\u91cd\u590d\u60e9\u7f5a<\/strong>\uff1a\u901a\u8fc7\u8c03\u6574\u6e29\u5ea6\u548c\u91cd\u590d\u60e9\u7f5a\u53c2\u6570\uff0c\u63a7\u5236\u751f\u6210\u8bed\u97f3\u7684\u591a\u6837\u6027\u548c\u6d41\u7545\u5ea6\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>\u5b89\u88c5 OuteTTS<\/strong>\uff1a\n<pre><code>pip install outetts\r\n<\/code><\/pre>\n<p>\u91cd\u8981\u63d0\u793a\uff1a\u5bf9\u4e8e GGUF \u652f\u6301\uff0c\u60a8\u9700\u8981\u5148\u624b\u52a8\u5b89\u88c5 <code>llama-cpp-python<\/code>\u3002\u8bf7\u8bbf\u95ee <a href=\"https:\/\/github.com\/abetlen\/llama-cpp-python\" target=\"_blank\" rel=\"noopener\">llama-cpp-python<\/a> \u83b7\u53d6\u5177\u4f53\u5b89\u88c5\u8bf4\u660e\u3002<\/li>\n<\/ol>\n<h3>\u4f7f\u7528\u65b9\u6cd5<\/h3>\n<ol>\n<li><strong>\u521d\u59cb\u5316\u63a5\u53e3<\/strong>\uff1a\n<pre><code>from outetts.v0_1.interface import InterfaceHF, InterfaceGGUF\r\n# \u4f7f\u7528 Hugging Face \u6a21\u578b\u521d\u59cb\u5316\u63a5\u53e3\r\ninterface = InterfaceHF(\"OuteAI\/OuteTTS-0.1-350M\")\r\n# \u6216\u8005\u4f7f\u7528 GGUF \u6a21\u578b\u521d\u59cb\u5316\u63a5\u53e3\r\n# interface = InterfaceGGUF(\"path\/to\/model.gguf\")\r\n<\/code><\/pre>\n<\/li>\n<li><strong>\u751f\u6210 TTS \u8f93\u51fa<\/strong>\uff1a\n<pre><code>output = interface.generate(\r\ntext=\"Hello, am I working?\",\r\ntemperature=0.1,\r\nrepetition_penalty=1.1,\r\nmax_length=4096\r\n)\r\n<\/code><\/pre>\n<\/li>\n<li><strong>\u64ad\u653e\u548c\u4fdd\u5b58\u751f\u6210\u7684\u97f3\u9891<\/strong>\uff1a\n<pre><code># \u64ad\u653e\u751f\u6210\u7684\u97f3\u9891\r\noutput.play()\r\n# \u4fdd\u5b58\u751f\u6210\u7684\u97f3\u9891\u5230\u6587\u4ef6\r\noutput.save(\"output.wav\")\r\n<\/code><\/pre>\n<\/li>\n<\/ol>\n<h3>\u8bed\u97f3\u514b\u9686<\/h3>\n<ol>\n<li><strong>\u521b\u5efa\u81ea\u5b9a\u4e49\u8bf4\u8bdd\u4eba<\/strong>\uff1a\n<pre><code>speaker = interface.create_speaker(\r\n\"path\/to\/reference.wav\",\r\n\"reference text matching the audio\"\r\n)\r\n<\/code><\/pre>\n<\/li>\n<li><strong>\u4fdd\u5b58\u548c\u52a0\u8f7d\u8bf4\u8bdd\u4eba<\/strong>\uff1a\n<pre><code># \u4fdd\u5b58\u8bf4\u8bdd\u4eba\u5230\u6587\u4ef6\r\ninterface.save_speaker(speaker, \"speaker.pkl\")\r\n# \u4ece\u6587\u4ef6\u52a0\u8f7d\u8bf4\u8bdd\u4eba\r\nspeaker = interface.load_speaker(\"speaker.pkl\")\r\n<\/code><\/pre>\n<\/li>\n<li><strong>\u4f7f\u7528\u81ea\u5b9a\u4e49\u8bed\u97f3\u751f\u6210 TTS<\/strong>\uff1a\n<pre><code>output = interface.generate(\r\ntext=\"This is a cloned voice speaking\",\r\nspeaker=speaker,\r\ntemperature=0.1,\r\nrepetition_penalty=1.1,\r\nmax_length=4096\r\n)\r\n<\/code><\/pre>\n<\/li>\n<\/ol>\n<h3>\u53c2\u6570\u8c03\u6574<\/h3>\n<ul>\n<li><strong>\u6e29\u5ea6\uff08temperature\uff09<\/strong>\uff1a\u63a7\u5236\u751f\u6210\u8bed\u97f3\u7684\u591a\u6837\u6027\u3002\u8f83\u4f4e\u7684\u6e29\u5ea6\uff08\u5982 0.1\uff09\u4f1a\u751f\u6210\u66f4\u786e\u5b9a\u7684\u8f93\u51fa\uff0c\u800c\u8f83\u9ad8\u7684\u6e29\u5ea6\uff08\u5982 0.7\uff09\u4f1a\u751f\u6210\u66f4\u591a\u6837\u5316\u7684\u8f93\u51fa\u3002<\/li>\n<li><strong>\u91cd\u590d\u60e9\u7f5a\uff08repetition_penalty\uff09<\/strong>\uff1a\u63a7\u5236\u751f\u6210\u8bed\u97f3\u7684\u91cd\u590d\u7a0b\u5ea6\u3002\u8f83\u9ad8\u7684\u91cd\u590d\u60e9\u7f5a\uff08\u5982 1.1\uff09\u53ef\u4ee5\u51cf\u5c11\u91cd\u590d\u5185\u5bb9\u7684\u751f\u6210\u3002<\/li>\n<\/ul>\n<p>\u901a\u8fc7\u4ee5\u4e0a\u6b65\u9aa4\uff0c\u7528\u6237\u53ef\u4ee5\u8f7b\u677e\u5b89\u88c5\u548c\u4f7f\u7528 OuteTTS \u6a21\u578b\u8fdb\u884c\u6587\u672c\u8f6c\u8bed\u97f3\u548c\u8bed\u97f3\u514b\u9686\u64cd\u4f5c\u3002\u8be6\u7ec6\u7684\u53c2\u6570\u8c03\u6574\u548c\u4f7f\u7528\u793a\u4f8b\u53ef\u4ee5\u5e2e\u52a9\u7528\u6237\u6839\u636e\u5177\u4f53\u9700\u6c42\u751f\u6210\u9ad8\u8d28\u91cf\u7684\u8bed\u97f3\u8f93\u51fa\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>OuteTTS \u662f\u4e00\u79cd\u5b9e\u9a8c\u6027\u6587\u672c\u8f6c\u8bed\u97f3\uff08TTS\uff09\u6a21\u578b\uff0c\u91c7\u7528\u7eaf\u8bed\u8a00\u5efa\u6a21\u65b9\u6cd5\u751f\u6210\u9ad8\u8d28\u91cf\u8bed\u97f3\u3002\u4e0e\u4f20\u7edf\u7684 TTS \u7cfb\u7edf\u4e0d\u540c\uff0cOuteTTS \u4e0d\u9700\u8981\u5916\u90e8\u9002\u914d\u5668\u6216\u590d\u6742\u7684\u67b6\u6784\u3002\u8be5\u6a21\u578b\u57fa\u4e8e LLaMa \u67b6\u6784\uff0c\u652f\u6301\u8bed\u97f3\u514b\u9686\u529f\u80fd\uff0c\u80fd\u591f\u751f\u6210\u5177\u6709\u968f\u673a\u8bf4\u8bdd\u4eba\u7279\u5f81\u7684\u8bed&#8230;<\/p>\n","protected":false},"author":1,"featured_media":61205,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20,392,400],"tags":[230,215],"class_list":["post-13082","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-tool","category-models","category-speech-model","tag-aikaiyuanxiangmu","tag-aiwenbenzhuanyuyin"],"_links":{"self":[{"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/posts\/13082","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=13082"}],"version-history":[{"count":0,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/posts\/13082\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/media\/61205"}],"wp:attachment":[{"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/media?parent=13082"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/categories?post=13082"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/tags?post=13082"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}