GithubHelp home page GithubHelp logo

time_dataset.py error about phasen HOT 5 CLOSED

huyanxin avatar huyanxin commented on June 19, 2024
time_dataset.py error

from phasen.

Comments (5)

huyanxin avatar huyanxin commented on June 19, 2024

You can do this to add duration information in tr.lst or cv.lst
python /tools/add_duration.py xxxx.lst
Sorry, I forgot adding it in ReadMe, and I will add it soon.

抱歉,我忘记在Readme里加这部分了,这个duration信息其实就是时长信息,在tr.lst和cv.lst加入duration后就可以在语音切片时能够预先拿到时长信息来切片;而语音切片的操作是因为语音长度不一,所以就把一个batch内语音切成等长的(padding策略亦可,但是速度会慢一些)。

from phasen.

LucySha avatar LucySha commented on June 19, 2024

@huyanxin 非常感谢,将duration加进去之后,代码可以运行起来。
还有几个问题想请教您:
1.time_dataset.py line 44行中,产生的lst每一个item包含干净和带噪音频的路径,分别对应inputs和labels吗?这个labels能用到哪些地方?
2. PHASEN的实测效果怎样?能否提供pretrained model,我之前对比了wavenet,segan以及deepfeatureloss,训练后模型进行实测与数据集的结果还是有差距的。

from phasen.

huyanxin avatar huyanxin commented on June 19, 2024
  1. 是这样,inputs就是带噪语音noisy-wav,labels就是clean-wav或者说是reference,给网络用来计算loss的,你看我代码https://github.com/huyanxin/phasen/blob/31ee2f1ba89b535142a5189abf913e9ac7f36404/steps/run_phasen.py#L74,在这里计算了Power Compress loss;
  2. 目前看的话,我这边训练的无论是Phasen还是其他模型,在实际样例上都还有能接受的效果吧,当然和训练集越匹配效果越好,Readme中的语谱图就是从实录样例中导出的,你想要模型的话把邮箱给我吧,我发你一个。

from phasen.

LucySha avatar LucySha commented on June 19, 2024

@huyanxin 邮箱[email protected] 感谢您的分享。我之前把模型用到实测环境时,总体降噪效果是可以接受的,但是会遇到比较明显的两个问题:1. 语音失真,表现在中文发音降噪后会出现断音的情况,2.用librispeech训练后的模型有的降噪后的音频会有明显的底噪,类似白噪声。请问您之前有碰到过类似的问题吗?处理过程中应该怎样尽可能避免呢?

from phasen.

huyanxin avatar huyanxin commented on June 19, 2024

OK,已发

from phasen.

Related Issues (10)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.