楼主: ReneeBK
1267 0

srvk_education [推广有奖]

  • 1关注
  • 62粉丝

VIP

学术权威

14%

还不是VIP/贵宾

-

TA的文库  其他...

R资源总汇

Panel Data Analysis

Experimental Design

威望
1
论坛币
49407 个
通用积分
51.8704
学术水平
370 点
热心指数
273 点
信用等级
335 点
经验
57815 点
帖子
4006
精华
21
在线时间
582 小时
注册时间
2005-5-8
最后登录
2023-11-26

相似文件 换一批

+2 论坛币
k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群
赵安豆老师微信:zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

感谢您参与论坛问题回答

经管之家送您两个论坛币!

+2 论坛币
srvk_education

Educational tutorials for speech and language processing classes

This repository contains materials that can be utilized in in-class tutorials for speech and language processing classes. The original material in this repository comes from CSE 5525: Foundations of Speech & Language Processing at The Ohio State University.

The material in this repository is geared towards advanced undergraduates and graduate students. Many of the tutorials assume a programming background, and at least a passing familiarity with python.

A virtual machine is used to house the resources; this VM can be built using Vagrant. After installing vagrant, issuing “vagrant up” in the directory where the Vagrantfile is will create a new VM with all of the materials.

Note that this version has some missing data that is commonly available in the speech community, which can be installed in a directory called "protected" parallel to the Vagarant file before initializing the vagrant vm. As of now, there are four protected data sources:

  • labeling_lab: contains sentences from TIMIT converted into audacity format. samp_sa1.aup contains one version of "She held your dark suit in greasy wash water all year" with labels. sa1.aup cotains another version of the same sentence from another speaker, but without labels (but word labels) for students to label. sa2.aup contains a different sentence.

  • lm_lab: contains the 5k language model data from WSJ0. Can easily be replaced with other free data.

  • srilm-1.7.1.tar.gz: contains a downloaded version of SRILM (which is free for instructional purposes but must be downloaded from SRI - web search for "srilm toolkit" to find the current version. You will need to change the Vagrantfile to reflect the current version.

  • tidigits.tgz: contains a copy of the TIDIGITS database, using the following ordering


bash-3.2$ tar -tzf protected/tidigits.tgz | headtidigits/tidigits/data/tidigits/data/adults/tidigits/data/adults/test/tidigits/data/adults/test/man/tidigits/data/adults/test/man/ah/tidigits/data/adults/test/man/ah/111a.wavtidigits/data/adults/test/man/ah/139oa.wavtidigits/data/adults/test/man/ah/155a.wavtidigits/data/adults/test/man/ah/1688a.wav

If you have licenses for WSJ, TIMIT or TIDIGITS and would like to know how to create the files needed for the experiments, please contact Eric Fosler-Lussier.

TO DO: rewrite tutorials using licensed data/toolkits with open data and toolkits.

本帖隐藏的内容

srvk_education-master.zip (1.33 MB)


二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

关键词:Education ducation cation ATION CATI background material original classes towards

您需要登录后才可以回帖 登录 | 我要注册

本版微信群
加好友,备注jltj
拉您入交流群

京ICP备16021002-2号 京B2-20170662号 京公网安备 11010802022788号 论坛法律顾问:王进律师 知识产权保护声明   免责及隐私声明

GMT+8, 2024-5-1 07:46