{"id":379,"date":"2025-05-05T23:38:09","date_gmt":"2025-05-05T15:38:09","guid":{"rendered":"https:\/\/www.crystal-cloud.top\/?p=379"},"modified":"2025-05-05T23:38:09","modified_gmt":"2025-05-05T15:38:09","slug":"%e4%bb%8e%e6%96%87%e6%9c%ac%e5%88%b0%e9%80%bc%e7%9c%9f%e8%af%ad%e9%9f%b3%ef%bc%9acoqui-tts%e5%b8%a6%e4%bd%a0%e8%bf%88%e5%85%a5ai%e9%85%8d%e9%9f%b3%e6%96%b0%e6%97%b6%e4%bb%a3","status":"publish","type":"post","link":"https:\/\/www.crystal-cloud.top\/index.php\/2025\/05\/05\/%e4%bb%8e%e6%96%87%e6%9c%ac%e5%88%b0%e9%80%bc%e7%9c%9f%e8%af%ad%e9%9f%b3%ef%bc%9acoqui-tts%e5%b8%a6%e4%bd%a0%e8%bf%88%e5%85%a5ai%e9%85%8d%e9%9f%b3%e6%96%b0%e6%97%b6%e4%bb%a3\/","title":{"rendered":"\u4ece\u6587\u672c\u5230\u903c\u771f\u8bed\u97f3\uff1aCoqui TTS\u5e26\u4f60\u8fc8\u5165AI\u914d\u97f3\u65b0\u65f6\u4ee3"},"content":{"rendered":"
\n

\u5728 AI \u8bed\u97f3\u5408\u6210\u9886\u57df\uff0cCoqui TTS \u4f5c\u4e3a\u4e00\u6b3e\u5f3a\u5927\u7684\u5f00\u6e90\u6587\u672c\u8f6c\u8bed\u97f3\uff08TTS\uff09\u5de5\u5177\uff0c\u6b63\u5728\u9010\u6b65\u6539\u53d8\u6211\u4eec\u7684\u5de5\u4f5c\u65b9\u5f0f\u3002\u5b83\u4e0d\u4ec5\u652f\u6301\u591a\u79cd\u9ad8\u8d28\u91cf\u8bed\u97f3\u6a21\u578b\uff0c\u8fd8\u5177\u5907\u58f0\u97f3\u514b\u9686\u3001\u591a\u8bed\u8a00\u652f\u6301\u3001\u4f4e\u5ef6\u8fdf\u63a8\u7406\u7b49\u529f\u80fd\uff0c\u8ba9\u5f00\u53d1\u8005\u53ef\u4ee5\u81ea\u7531\u6784\u5efa\u5b9a\u5236\u5316\u7684\u8bed\u97f3\u5408\u6210\u7cfb\u7edf\u3002<\/span><\/p>\n

Coqui TTS \u4ecb\u7ecd<\/span><\/strong><\/h2>\n

Coqui TTS \u6e90\u4e8e Mozilla TTS \u9879\u76ee\uff0c\u7ecf\u8fc7\u4e0d\u65ad\u8fed\u4ee3\u4f18\u5316\uff0c\u5df2\u6210\u4e3a\u72ec\u7acb\u4e14\u6210\u719f\u7684 AI \u8bed\u97f3\u5408\u6210\u6846\u67b6\u3002\u5b83\u91c7\u7528\u591a\u79cd\u6700\u5148\u8fdb\u7684 TTS \u6a21\u578b\uff0c\u5982 Tacotron2\u3001Glow-TTS\u3001VITS \u7b49\uff0c\u80fd\u751f\u6210\u9ad8\u5ea6\u81ea\u7136\u7684\u8bed\u97f3\uff0c\u751a\u81f3\u652f\u6301\u7279\u5b9a\u97f3\u8272\u514b\u9686\uff0c\u9002\u7528\u4e8e\u5404\u79cd\u573a\u666f\uff0c\u5982\u8bed\u97f3\u52a9\u624b\u3001\u667a\u80fd\u5ba2\u670d\u3001\u64ad\u5ba2\u914d\u97f3\u7b49\u3002<\/span><\/p>\n

\"\u56fe\u7247\"<\/section>\n

\u6838\u5fc3\u529f\u80fd<\/span><\/strong><\/h2>\n

1. \u9ad8\u8d28\u91cf\u8bed\u97f3\u5408\u6210<\/span><\/strong><\/h3>\n

Coqui TTS \u63d0\u4f9b\u591a\u79cd\u6587\u672c\u8f6c\u8bed\u97f3\u6a21\u578b\uff0c\u53ef\u751f\u6210\u81ea\u7136\u6d41\u7545\u7684\u8bed\u97f3\uff0c\u51cf\u5c11\u7535\u5b50\u5408\u6210\u97f3\u7684\u751f\u786c\u611f\uff0c\u63a5\u8fd1\u771f\u4eba\u8bed\u97f3\u6548\u679c\u3002<\/span><\/p>\n

2. \u591a\u8bed\u8a00\u652f\u6301<\/span><\/strong><\/h3>\n

\u652f\u6301\u591a\u79cd\u8bed\u8a00\uff0c\u5305\u62ec\u4e2d\u6587\u3001\u82f1\u6587\u3001\u65e5\u8bed\u3001\u6cd5\u8bed\u3001\u897f\u73ed\u7259\u8bed\u7b49\uff0c\u53ef\u4ee5\u6ee1\u8db3\u4e0d\u540c\u8bed\u8a00\u73af\u5883\u4e0b\u7684\u8bed\u97f3\u5408\u6210\u9700\u6c42\u3002<\/span><\/p>\n

3. \u8bed\u97f3\u514b\u9686\u4e0e\u4e2a\u6027\u5316\u58f0\u97f3\u5b9a\u5236<\/span><\/strong><\/h3>\n

\u901a\u8fc7 YourTTS \u6a21\u578b\uff0c\u7528\u6237\u53ef\u4ee5\u7528\u5c11\u91cf\u97f3\u9891\u6837\u672c\u514b\u9686\u81ea\u5df1\u7684\u58f0\u97f3\uff0c\u751f\u6210\u4e2a\u6027\u5316 TTS \u6a21\u578b\uff0c\u9002\u7528\u4e8e\u914d\u97f3\u3001\u97f3\u9891\u5185\u5bb9\u521b\u4f5c\u7b49\u9886\u57df\u3002<\/span><\/p>\n

4. \u4f4e\u5ef6\u8fdf\u63a8\u7406\uff0c\u5b9e\u65f6\u751f\u6210\u8bed\u97f3<\/span><\/strong><\/h3>\n

Coqui TTS \u63d0\u4f9b\u8f7b\u91cf\u7ea7\u4e0e\u9ad8\u8d28\u91cf\u6a21\u578b\u9009\u62e9\uff0c\u5728\u4fdd\u8bc1\u97f3\u8d28\u7684\u540c\u65f6\uff0c\u5b9e\u73b0\u4f4e\u5ef6\u8fdf\u63a8\u7406\uff0c\u9002\u7528\u4e8e\u76f4\u64ad\u3001\u667a\u80fd\u8bed\u97f3\u52a9\u624b\u7b49\u5b9e\u65f6\u5e94\u7528\u573a\u666f\u3002<\/span><\/p>\n

5. \u7075\u6d3b\u6269\u5c55\uff0c\u652f\u6301\u591a\u79cd TTS \u67b6\u6784<\/span><\/strong><\/h3>\n

Coqui TTS \u91c7\u7528\u6a21\u5757\u5316\u8bbe\u8ba1\uff0c\u652f\u6301 Tacotron2\u3001Glow-TTS\u3001VITS\u3001SpeedySpeech \u7b49\u591a\u79cd TTS \u6a21\u578b\uff0c\u65b9\u4fbf\u5f00\u53d1\u8005\u6839\u636e\u9700\u6c42\u9009\u62e9\u6700\u4f18\u65b9\u6848\u3002<\/span><\/p>\n

\"\u56fe\u7247\"<\/section>\n

\u5b89\u88c5\u4e0e\u4f7f\u7528<\/span><\/strong><\/h2>\n

1. \u5b89\u88c5 Coqui TTS<\/span><\/strong><\/h3>\n

\u53ef\u4ee5\u76f4\u63a5\u4f7f\u7528 pip \u8fdb\u884c\u5b89\u88c5\uff1a<\/span><\/p>\n

<\/span><\/span><\/span><\/span> <\/span><\/span><\/section>
pip install <\/span>TTS<\/span><\/span>
<\/span><\/code><\/pre>\n
\u5982\u679c\u5e0c\u671b\u5b89\u88c5\u6700\u65b0\u5f00\u53d1\u7248\u672c\uff0c\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u547d\u4ee4\uff1a<\/span><\/p>\n
<\/span><\/span><\/span><\/span> <\/span><\/span><\/section>
pip install git<\/span>+<\/span><\/span>https<\/span>:<\/span><\/span>\/<\/span><\/span>\/<\/span><\/span>github<\/span>.<\/span><\/span>com<\/span><\/span>\/<\/span><\/span>coqui<\/span>-<\/span><\/span>ai<\/span>\/<\/span><\/span>TTS<\/span><\/span>
<\/span><\/code><\/pre>\n
\u60f3\u8981\u5b8c\u6574\u5b89\u88c5\u6240\u6709\u529f\u80fd\uff08\u5305\u62ec\u58f0\u97f3\u514b\u9686\u3001\u591a\u8bf4\u8bdd\u4eba\u652f\u6301\u7b49\uff09\uff0c\u53ef\u4ee5\u4f7f\u7528\uff1a<\/span><\/p>\n
<\/span><\/span><\/span><\/span> <\/span><\/span><\/section>
pip install <\/span>TTS<\/span><\/span>[<\/span><\/span>all<\/span>]<\/span><\/span>
<\/span><\/code><\/pre>\n
Docker \u955c\u50cf<\/span><\/p>\n
\u60a8\u4e5f\u53ef\u4ee5\u4f7f\u7528 docker \u955c\u50cf\u5c1d\u8bd5\u65e0\u9700\u5b89\u88c5\u7684 TTS\u3002\u53ea\u9700\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4\uff0c\u60a8\u65e0\u9700\u5b89\u88c5\u5373\u53ef\u8fd0\u884c TTS\u3002<\/span><\/p>\n
<\/span><\/span><\/span><\/span> <\/span><\/span><\/section>
docker<\/span><\/span> run <\/span>--rm<\/span><\/span> <\/span>-it<\/span><\/span> <\/span>-p<\/span><\/span> <\/span>5002<\/span><\/span>:5002 <\/span>--entrypoint<\/span><\/span> \/bin\/bash ghcr.io\/coqui-ai\/tts-cpu<\/span>
<\/span>python3 TTS\/server\/server.py <\/span>--list_models<\/span><\/span> <\/span>#To get the list of available models<\/span><\/span>
<\/span>python3 TTS\/server\/server.py <\/span>--model_name<\/span><\/span> tts_models\/en\/vctk\/vits <\/span># To start a server<\/span><\/span>
<\/span>
<\/span>\u5b89\u88c5\u540e\u8bbf\u95eehttp:\/\/<\/span>[<\/span><\/span>::1<\/span>]<\/span><\/span>:5002\/<\/span>
<\/span><\/code><\/pre>\n
\"\u56fe\u7247\"<\/section>\n
2. \u8fd0\u884c TTS \u8fdb\u884c\u8bed\u97f3\u5408\u6210<\/span><\/strong><\/h3>\n
\u5b89\u88c5\u5b8c\u6210\u540e\uff0c\u53ef\u4ee5\u76f4\u63a5\u5728\u547d\u4ee4\u884c\u4e2d\u8fd0\u884c TTS \u8fdb\u884c\u8bed\u97f3\u5408\u6210\uff1a<\/span><\/p>\n
<\/span><\/span><\/span><\/span> <\/span><\/span><\/section>
tts <\/span>--text<\/span><\/span> <\/span>\"\u4f60\u597d\uff0c\u8fd9\u662f\u4e00\u4e2a\u6d4b\u8bd5\"<\/span><\/span> <\/span>--model_name<\/span><\/span> <\/span>\"tts_models\/zh-CN\/baker\/tacotron2-DDC-GST\"<\/span><\/span> <\/span>--out_path<\/span><\/span> output.wav<\/span>
<\/span><\/code><\/pre>\n
\u5982\u679c\u5e0c\u671b\u5728 Python \u4ee3\u7801\u4e2d\u8c03\u7528 Coqui TTS\uff0c\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u65b9\u5f0f\uff1a<\/span><\/p>\n
<\/span><\/span><\/span><\/span> <\/span><\/span><\/section>
from TTS.api <\/span>import<\/span><\/span> TTS<\/span>
<\/span>
<\/span>tts <\/span>=<\/span><\/span> TTS<\/span>(<\/span><\/span>model_name<\/span>=<\/span><\/span>\"tts_models\/zh-CN\/baker\/tacotron2-DDC-GST\"<\/span><\/span>)<\/span><\/span>
<\/span>tts.tts_to_file<\/span>(<\/span><\/span>text<\/span>=<\/span><\/span>\"\u4f60\u597d\uff0c\u8fd9\u662f\u4e00\u4e2a\u6d4b\u8bd5\"<\/span><\/span>, <\/span>file_path<\/span><\/span>=<\/span><\/span>\"output.wav\"<\/span><\/span>)<\/span><\/span>
<\/span><\/code><\/pre>\n
3. \u8bed\u97f3\u514b\u9686<\/span><\/strong><\/h3>\n
\u5bf9\u4e8e YourTTS \u7b49\u652f\u6301\u591a\u8bf4\u8bdd\u4eba\u548c\u8bed\u97f3\u514b\u9686\u7684\u6a21\u578b\uff0c\u53ef\u4ee5\u6307\u5b9a\u8bf4\u8bdd\u4eba ID \u8fdb\u884c\u4e2a\u6027\u5316\u5408\u6210\uff1a<\/span><\/p>\n
<\/span><\/span><\/span><\/span> <\/span><\/span><\/section>
tts <\/span>=<\/span><\/span> TTS<\/span>(<\/span><\/span>model_name<\/span>=<\/span><\/span>\"tts_models\/multilingual\/multi-dataset\/your_tts\"<\/span><\/span>)<\/span><\/span>
<\/span>tts.tts_to_file<\/span>(<\/span><\/span>text<\/span>=<\/span><\/span>\"Hello world\"<\/span><\/span>, <\/span>file_path<\/span><\/span>=<\/span><\/span>\"output.wav\"<\/span><\/span>, <\/span>speaker<\/span><\/span>=<\/span><\/span>\"p326\"<\/span><\/span>)<\/span><\/span>
<\/span><\/code><\/pre>\n
4. \u8bad\u7ec3\u81ea\u5b9a\u4e49\u8bed\u97f3\u6a21\u578b<\/span><\/strong><\/h3>\n
\u5982\u679c\u60f3\u8981\u8bad\u7ec3\u81ea\u5df1\u7684\u8bed\u97f3\u6a21\u578b\uff0c\u53ef\u4ee5\u51c6\u5907\u6570\u636e\u96c6\uff08\u97f3\u9891\u6587\u4ef6+\u6587\u672c\uff09\u5e76\u8fdb\u884c\u8bad\u7ec3\u3002\u4f8b\u5982\uff0c\u4f7f\u7528 LJSpeech \u683c\u5f0f\u6570\u636e\u96c6\uff1a<\/span><\/p>\n
<\/span><\/span><\/span><\/span> <\/span><\/span><\/section>
tts train <\/span>--config_path<\/span><\/span> config.json <\/span>--coqpit.output_path<\/span><\/span> .\/training_output<\/span>
<\/span><\/code><\/pre>\n
\u5982\u679c\u5df2\u6709\u9884\u8bad\u7ec3\u6a21\u578b\uff0c\u53ef\u4ee5\u57fa\u4e8e\u5df2\u6709\u6a21\u578b\u8fdb\u884c\u5fae\u8c03\uff08Fine-tune\uff09\uff1a<\/span><\/p>\n
<\/span><\/span><\/span><\/span> <\/span><\/span><\/section>
tts train <\/span>--config_path<\/span><\/span> config.json <\/span>--coqpit.output_path<\/span><\/span> .\/finetune_output <\/span>--restore_path<\/span><\/span> path\/to\/pretrained_model.pth<\/span>
<\/span><\/code><\/pre>\n
\u4f7f\u7528\u573a\u666f<\/span><\/strong><\/h2>\n
    \n
  1. AI \u8bed\u97f3\u52a9\u624b<\/span><\/strong>
    \n
    \uff1a\u642d\u5efa\u5177\u5907\u4e2a\u6027\u5316\u58f0\u97f3\u7684\u8bed\u97f3\u52a9\u624b\uff0c\u8ba9 AI \u66f4\u5177\u4eba\u6027\u5316\u3002<\/span><\/section>\n<\/li>\n
  2. \u667a\u80fd\u5ba2\u670d\u7cfb\u7edf<\/span><\/strong>
    \n
    \uff1a\u7528\u4e8e\u4f01\u4e1a\u7684\u81ea\u52a8\u8bed\u97f3\u5e94\u7b54\u7cfb\u7edf\uff0c\u63d0\u9ad8\u5ba2\u670d\u6548\u7387\u3002<\/span><\/section>\n<\/li>\n
  3. \u64ad\u5ba2\u4e0e\u6709\u58f0\u8bfb\u7269<\/span><\/strong>
    \n
    \uff1a\u5c06\u6587\u672c\u5185\u5bb9\u8f6c\u5316\u4e3a\u81ea\u7136\u8bed\u97f3\uff0c\u52a9\u529b\u5185\u5bb9\u521b\u4f5c\u3002<\/span><\/section>\n<\/li>\n
  4. \u6559\u80b2\u4e0e\u8bed\u97f3\u8bad\u7ec3<\/span><\/strong>
    \n
    \uff1a\u5e2e\u52a9\u8bed\u8a00\u5b66\u4e60\u8005\u7ec3\u4e60\u53d1\u97f3\uff0c\u63d0\u4f9b\u8bed\u97f3\u53cd\u9988\u3002<\/span><\/section>\n<\/li>\n
  5. \u77ed\u89c6\u9891\u914d\u97f3<\/span><\/strong>
    \n
    \uff1a\u81ea\u52a8\u751f\u6210\u8bed\u97f3\u5185\u5bb9\uff0c\u4e3a\u89c6\u9891\u521b\u4f5c\u63d0\u4f9b\u914d\u97f3\u65b9\u6848\u3002<\/span><\/section>\n<\/li>\n<\/ol>\n
    \u603b\u7ed3<\/span><\/strong><\/h2>\n
    Coqui TTS \u4f5c\u4e3a\u5f00\u6e90\u6587\u672c\u8f6c\u8bed\u97f3\u5de5\u5177\uff0c\u5728\u8bed\u97f3\u5408\u6210\u8d28\u91cf\u3001\u591a\u8bed\u8a00\u652f\u6301\u3001\u4e2a\u6027\u5316\u8bed\u97f3\u514b\u9686\u7b49\u65b9\u9762\u8868\u73b0\u4f18\u5f02\uff0c\u9002\u7528\u4e8e\u5404\u79cd\u8bed\u97f3\u751f\u6210\u9700\u6c42\u3002\u65e0\u8bba\u662f\u5f00\u53d1 AI \u8bed\u97f3\u52a9\u624b\uff0c\u8fd8\u662f\u8fdb\u884c\u4e2a\u6027\u5316\u8bed\u97f3\u5b9a\u5236\uff0cCoqui TTS \u90fd\u662f\u4e00\u4e2a\u503c\u5f97\u63a8\u8350\u7684\u89e3\u51b3\u65b9\u6848\u3002<\/span><\/p>\n<\/section>\n

    <\/span><\/section>\n
    <\/mp-style-type><\/p>\n","protected":false},"excerpt":{"rendered":"
    \u5728 AI \u8bed\u97f3\u5408\u6210\u9886\u57df\uff0cCoqui TTS \u4f5c\u4e3a\u4e00\u6b3e\u5f3a\u5927\u7684\u5f00\u6e90\u6587\u672c\u8f6c\u8bed\u97f3\uff08TTS\uff09\u5de5\u5177\uff0c\u6b63\u5728\u9010\u6b65\u6539\u53d8\u6211\u4eec\u7684\u5de5\u4f5c […]<\/p>\n","protected":false},"author":1,"featured_media":384,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[10],"tags":[],"class_list":["post-379","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai"],"_links":{"self":[{"href":"https:\/\/www.crystal-cloud.top\/index.php\/wp-json\/wp\/v2\/posts\/379","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.crystal-cloud.top\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.crystal-cloud.top\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.crystal-cloud.top\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.crystal-cloud.top\/index.php\/wp-json\/wp\/v2\/comments?post=379"}],"version-history":[{"count":2,"href":"https:\/\/www.crystal-cloud.top\/index.php\/wp-json\/wp\/v2\/posts\/379\/revisions"}],"predecessor-version":[{"id":385,"href":"https:\/\/www.crystal-cloud.top\/index.php\/wp-json\/wp\/v2\/posts\/379\/revisions\/385"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.crystal-cloud.top\/index.php\/wp-json\/wp\/v2\/media\/384"}],"wp:attachment":[{"href":"https:\/\/www.crystal-cloud.top\/index.php\/wp-json\/wp\/v2\/media?parent=379"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.crystal-cloud.top\/index.php\/wp-json\/wp\/v2\/categories?post=379"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.crystal-cloud.top\/index.php\/wp-json\/wp\/v2\/tags?post=379"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}