Building Corpora for Single-Channel Speech Separation Across Multiple Domains
To date, the bulk of research on single-channel speech separation has been conducted using clean, near-field, read speech, which is not representative of many modern applications. In this work, we develop a procedure for constructing high-quality synthetic overlap datasets, necessary for most deep learning-based separation frameworks. We produced datasets that are more representative of realistic applications using the CHiME-5 and Mixer 6 corpora and evaluate standard methods on this data to demonstrate the shortcomings of current source-separation performance. We also demonstrate the value of a wide variety of data in training robust models that generalize well to multiple conditions.
Authors

Are you an author of this paper? Check the Twitter handle we have for you is correct.

Matthew Maciejewski (add twitter)
Gregory Sell (edit)
Leibny Paola Garcia-Perera (add twitter)
Shinji Watanabe (add twitter)
Sanjeev Khudanpur (add twitter)
Ask The Authors

Ask the authors of this paper a question or leave a comment.

Read it. Rate it.
#1. Which part of the paper did you read?

#2. The paper contains new data or analyses that is openly accessible?
#3. The conclusion is supported by the data and analyses?
#4. The conclusion is of scientific interest?
#5. The result is likely to lead to future research?

Github
Repo:
Stargazers:
11
Forks:
5
Open Issues:
0
Network:
2399
Subscribers:
2
Language:
Shell
This is now the official location of the Kaldi project.
Youtube
Link:
None (add)
Views:
0
Likes:
0
Dislikes:
0
Favorites:
0
Comments:
0
Other
Sample Sizes (N=):
Inserted:
Words Total:
Words Unique:
Source:
Abstract:
None
11/07/18 06:05PM
4,014
1,464
Tweets
arxivml: "Building Corpora for Single-Channel Speech Separation Across Multiple Domains", Matthew Maciejewski, Gregory Sell,… https://t.co/OGmhlJeZOC
arxiv_cscl: Building Corpora for Single-Channel Speech Separation Across Multiple Domains https://t.co/VcP8yUA8Ds
nmfeeds: [O] https://t.co/w2InlunNHG Building Corpora for Single-Channel Speech Separation Across Multiple Domains. To date, the bu...
nmfeeds: [CL] https://t.co/w2InlunNHG Building Corpora for Single-Channel Speech Separation Across Multiple Domains. To date, the b...
arxiv_cscl: Building Corpora for Single-Channel Speech Separation Across Multiple Domains https://t.co/VcP8yUA8Ds
BrundageBot: Building Corpora for Single-Channel Speech Separation Across Multiple Domains. Matthew Maciejewski, Gregory Sell, Leibny Paola Garcia-Perera, Shinji Watanabe, and Sanjeev Khudanpur https://t.co/2OI1kkxatb
arxiv_cscl: Building Corpora for Single-Channel Speech Separation Across Multiple Domains https://t.co/VcP8yUA8Ds
Images
Related