Difference between revisions of "Publications/riols.16.seminar"
From LRDE
Fanny Riols (talk | contribs) |
|||
| Line 1: | Line 1: | ||
{{CSIReport |
{{CSIReport |
||
| authors = Fanny Riols |
| authors = Fanny Riols |
||
| − | | title = Mel Frequency Cepstral Coefficients |
+ | | title = Speaker Diarization based on the Mel Frequency Cepstral Coefficients |
| year = 2016 |
| year = 2016 |
||
| number = 1515 |
| number = 1515 |
||
| − | | abstract = Speaker diarization has emerged as an increasingly important and dedicated domain of speech research. It relates to the problem of determining "who spoke when?". It means that we would like to find the intervals during which |
+ | | abstract = Speaker diarization has emerged as an increasingly important and dedicated domain of speech research. It relates to the problem of determining "who spoke when ?". It means that we would like to find the intervals during which eachspeaker is active. By computing the Mel Frequency Cepstral Coefficients (MFCC) features from a given speech signal and using the Independent Component Analysis (ICA) on these features, we are able to segment the speech, with the help of a Hidden Markov Model (HMM). We will use this algorithm for speaker diarization in verification systemwith multi-speaker audio data, such as interview of microphone segment of NIST Speaker Recognition Evaluation. |
| type = techreport |
| type = techreport |
||
| id = riols.16.seminar |
| id = riols.16.seminar |
||
| − | | bibtex = |
||
| − | @TechReport<nowiki>{</nowiki> riols.16.seminar, |
||
| − | author = <nowiki>{</nowiki>Fanny Riols<nowiki>}</nowiki>, |
||
| − | title = <nowiki>{</nowiki>Mel Frequency Cepstral Coefficients based Speaker |
||
| − | Diarization<nowiki>}</nowiki>, |
||
| − | institution = <nowiki>{</nowiki>LRDE<nowiki>}</nowiki>, |
||
| − | year = 2016, |
||
| − | address = <nowiki>{</nowiki>Paris, France<nowiki>}</nowiki>, |
||
| − | month = jan, |
||
| − | abstract = <nowiki>{</nowiki> Speaker diarization has emerged as an increasingly |
||
| − | important and dedicated domain of speech research. It relates to the |
||
| − | problem of determining "who spoke when?". It means that we would like to |
||
| − | find the intervals during which each speaker is active. By computing the |
||
| − | Mel Frequency Cepstral Coefficients (MFCC) features from a given speech |
||
| − | signal and using the Independent Component Analysis (ICA) on these |
||
| − | features, we are able to segment the speech with the help of a Hidden |
||
| − | Markov Model (HMM) and Gaussian Mixture Model (GMM). We will use this |
||
| − | algorithm for speaker diarization in verification system, with |
||
| − | multi-speaker audio data, such as interview of microphone segment of |
||
| − | NIST Speaker Recognition Evaluation.<nowiki>}</nowiki> |
||
| − | <nowiki>}</nowiki> |
||
| − | |||
}} |
}} |
||
Latest revision as of 06:19, 5 August 2018
- Authors
- Fanny Riols
- Type
- techreport
- Year
- 2016
- Number
- 1515
Abstract
Speaker diarization has emerged as an increasingly important and dedicated domain of speech research. It relates to the problem of determining "who spoke when ?". It means that we would like to find the intervals during which eachspeaker is active. By computing the Mel Frequency Cepstral Coefficients (MFCC) features from a given speech signal and using the Independent Component Analysis (ICA) on these features, we are able to segment the speech, with the help of a Hidden Markov Model (HMM). We will use this algorithm for speaker diarization in verification systemwith multi-speaker audio data, such as interview of microphone segment of NIST Speaker Recognition Evaluation.