Replies: 3 comments 1 reply
-
模型只会在第一次调用的时候下载😊 |
Beta Was this translation helpful? Give feedback.
0 replies
-
谢谢回复!前面写得不准确,是第一次会下载模型,每次都会加载模型。按#1177 ,Wenet只调用一次模型,paddlespeech慢4倍。请问有没有从本机调用模型、且只调用一次的ASR、TTS的Python代码呢? |
Beta Was this translation helpful? Give feedback.
0 replies
-
类似的功能在开发中了 |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
有好几个讨论里面提到识别和合成速度慢的问题(#1256; #1280),问题主要是现有的CL每次都重新下载模型;已有的Python代码“ASRExecutor”(https://github.com/PaddlePaddle/PaddleSpeech/tree/develop/demos/speech_recognition)、“TTSExecutor”(https://github.com/PaddlePaddle/PaddleSpeech/tree/develop/demos/text_to_speech)也会每次重新下载模型。如果将模型下载下来,请问是否有从本机调用模型、且只调用一次的ASR、TTS的Python代码呢?
Beta Was this translation helpful? Give feedback.
All reactions