search

PaddlePaddle / PaddleSpeech

Easy-to-use Speech Toolkit including Self-Supervised Learning model, SOTA/Streaming ASR with punctuation, Streaming TTS with text frontend, Speaker Verification System, End-to-End Speech Translation and Keyword Spotting. Won NAACL2022 Best Demo Award.
https://paddlespeech.readthedocs.io
Apache License 2.0
11.16k stars 1.85k forks source link

decoders/swig的 setup.sh下载 kenlm的git地址没了吗 #465

Closed lgd123 closed 4 years ago

lgd123 commented 4 years ago

https://github.com/luotao1/kenlm.git 我在家里访问这个地址 一直是404.。。。怎么办 本人新手，是作者删除了 还是我被墙了

zhupengyang commented 4 years ago

这是是个人的repo，可能被删了。这个问题我们尽快确认下

lgd123 commented 4 years ago

@zhupengyang 谢谢 大哥，我是新手 安装了3天没安装好...我用的是vmware虚拟机 安装的Ubuntu，安装出现了各种问题，swig出问题，安装openFst出问题，主要是openFst安装不上。我刚接触Python 我想问问， 1.vmware环境下 可以运行无gpu本项目吗。。 2.vmware环境下 无gpu,可以用Docker那个版本运行DeepSpeech吗 没Docker，手头没有真实linux机器 运行 DeepSpeech有点难。。呜呜

zhupengyang commented 4 years ago

1.vmware环境下 可以运行无gpu本项目吗。。 2.vmware环境下 无gpu,可以用Docker那个版本运行DeepSpeech吗

理论上都是可以的~

Aionrichman commented 4 years ago

@zhupengyang 请问这个git什么时候能好啊

rinjac commented 4 years ago

@zhupengyang @Aionrichman I fixed this issue by referencing the original KenLM repository (https://github.com/kpu/kenlm) instead of the unavailable one. For what I am using it, it works, but the creators need to confirm if the now unavailable repo (https://github.com/luotao1/kenlm.git) contained something which is not present in the original KenLM repo.

I created a pull request #464 which contains my fix.

yeyupiaoling commented 4 years ago

@Aionrichman @lgd123 已经修复啦，看新脚本代码， https://github.com/PaddlePaddle/DeepSpeech/blob/b3c728d46ff4eee68c45f20b0abb76e968008bcb/decoders/swig/setup.sh#L3-L9