%Aigaion2 BibTeX export from Idiap Publications
%Thursday 21 November 2024 12:51:06 PM

@INPROCEEDINGS{ajmera2001art,
         author = {Ajmera, Jitendra and McCowan, Iain A. and Bourlard, Herv{\'{e}}},
       projects = {Idiap},
          title = {Robust HMM-Based Speech/Music Segmentation},
      booktitle = {ICASSP},
           year = {2002},
        address = {Orlando, Florida},
           note = {IDIAP-RR 01-33},
       crossref = {ajmera-rr-01-33},
       abstract = {In this paper we present a new approach towards high performance speech/music segmentation on realistic tasks related to the automatic transcription of broadcast news. In the approach presented here, the local probability density function (PDF) estimators trained on clean microphone speech are used as a channel model at the output of which the entropy and ``dynamism'' will be measured and integrated over time through a 2-state (speech and and non-speech) hidden Markov model (HMM) with minimum duration constraints. The parameters of the HMM are trained using the EM algorithm in a completely unsupervised manner. Different experiments, including a variety of speech and music styles, as well as different segment durations of speech and music signals (real data distribution, mostly speech, or mostly music,',','),
 will illustrate the robustness of the approach, which in each case achieves a frame-level accuracy greater than 94\\%.},
            pdf = {https://publications.idiap.ch/attachments/reports/2001/ajmera2002icassp.pdf},
     postscript = {ftp://ftp.idiap.ch/pub/reports/2001/ajmera2002icassp.ps},
ipdmembership={speech},
}



crossreferenced publications: 
@TECHREPORT{ajmera-rr-01-33,
         author = {Ajmera, Jitendra and McCowan, Iain A. and Bourlard, Herv{\'{e}}},
       projects = {Idiap},
          title = {Robust HMM-Based Speech/Music Segmentation},
           type = {Idiap-RR},
         number = {Idiap-RR-33-2001},
           year = {2001},
    institution = {IDIAP},
        address = {Martigny, Switzerland},
           note = {ICASSP,Orlando, Florida, 2002},
       abstract = {In this paper we present a new approach towards high performance speech/music segmentation on realistic tasks related to the automatic transcription of broadcast news. In the approach presented here, the local probability density function (PDF) estimators trained on clean microphone speech are used as a channel model at the output of which the entropy and ``dynamism'' will be measured and integrated over time through a 2-state (speech and and non-speech) hidden Markov model (HMM) with minimum duration constraints. The parameters of the HMM are trained using the EM algorithm in a completely unsupervised manner. Different experiments, including a variety of speech and music styles, as well as different segment durations of speech and music signals (real data distribution, mostly speech, or mostly music,',','),
 will illustrate the robustness of the approach, which in each case achieves a frame-level accuracy greater than 94\\%.},
            pdf = {https://publications.idiap.ch/attachments/reports/2001/rr01-33.pdf},
     postscript = {ftp://ftp.idiap.ch/pub/reports/2001/rr01-33.ps.gz},
ipdinar={2001},
ipdmembership={speech},
language={English},
}