Skip to main content
×
Home
    • Aa
    • Aa

Deep-neural network approaches for speech recognition with heterogeneous groups of speakers including children

  • ROMAIN SERIZEL (a1) (a2) and DIEGO GIULIANI (a2)
Abstract
Abstract

This paper introduces deep neural network (DNN)–hidden Markov model (HMM)-based methods to tackle speech recognition in heterogeneous groups of speakers including children. We target three speaker groups consisting of children, adult males and adult females. Two different kind of approaches are introduced here: approaches based on DNN adaptation and approaches relying on vocal-tract length normalisation (VTLN). First, the recent approach that consists in adapting a general DNN to domain/language specific data is extended to target age/gender groups in the context of DNN–HMM. Then, VTLN is investigated by training a DNN–HMM system by using either mel frequency cepstral coefficients normalised with standard VTLN or mel frequency cepstral coefficients derived acoustic features combined with the posterior probabilities of the VTLN warping factors. In this later, novel, approach the posterior probabilities of the warping factors are obtained with a separate DNN and the decoding can be operated in a single pass when the VTLN approach requires two decoding passes. Finally, the different approaches presented here are combined to take advantage of their complementarity. The combination of several approaches is shown to improve the baseline phone error rate performance by thirty per cent to thirty-five per cent relative and the baseline word error rate performance by about ten per cent relative.

  • View HTML
    • Send article to Kindle

      To send this article to your Kindle, first ensure coreplatform@cambridge.org is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle.

      Note you can select to send to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

      Find out more about the Kindle Personal Document Service.

      Deep-neural network approaches for speech recognition with heterogeneous groups of speakers including children
      Available formats
      ×
      Send article to Dropbox

      To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your Dropbox account. Find out more about sending content to Dropbox.

      Deep-neural network approaches for speech recognition with heterogeneous groups of speakers including children
      Available formats
      ×
      Send article to Google Drive

      To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your Google Drive account. Find out more about sending content to Google Drive.

      Deep-neural network approaches for speech recognition with heterogeneous groups of speakers including children
      Available formats
      ×
Copyright
Footnotes
Hide All

This work was partially funded by the European project EU-BRIDGE, under the contract FP7-287658.

Footnotes
Linked references
Hide All

This list contains references from the content that can be linked to their source. For a full set of references and notes please see the PDF or HTML where available.

Y. Bengio , A. Courville , and P. Vincent , 2013. Representation learning: a review and new perspectives. IEEE Transactions on Pattern Analysis and Machine Intelligence 35 (8): 1798–828.

H. Bourlard , and N. Morgan 1994. Connectionist Speech Recognition: A Hybrid Approach, Springer, Berlin, Germany, vol. 247, Springer.

T. Claes , I. Dologlou , L. ten Bosch , and D. Van Compernolle 1998. A novel feature transformation for vocal tract length normalisation in automatic speech recognition. IEEE Transactions on Speech and Audio Processing 6 (6): 549–57.

G. E. Dahl , D. Yu , L. Deng , and A. Acero , 2012. Context-dependent pre-trained deep neural networks for large-vocabulary speech recognition. IEEE Transactions on Audio, Speech and Language Processing 20 (1): 3042.

N. Dehak , P. Kenny , R. Dehak , P. Dumouchel , and P. Ouellet , 2011. Front-end factor analysis for speaker verification. IEEE Transactions on Audio, Speech and Language Processing 19 (4): 788–98.

W. T. Fitch , and J. Giedd , 1999. Morphology and development of the human vocal tract: a study using magnetic resonance imaging. Journal of Acoustical Society of America 106 (3): 1511–22.

M. J. F. Gales , 1998. Maximum likelihood linear transformations for HMM-based speech recognition. Computer Speech and Language 12 (2): 7598.

M. Gerosa , D. Giuliani , and F. F. Brugnara , 2007. Acoustic variability and automatic recognition of children's speech. Speech Communication 49 (10–11): 847–60.

M. Gerosa , D. Giuliani , and F. Brugnara , 2009a. Towards age-independent acoustic modeling. Speech Communication 51 (6): 499509.

M. Gerosa , D. Giuliani , S. Narayanan , and A. Potamianos , 2009b. A Review of ASR technologies for Children's speech. In Proceedings of the 2nd Workshop on Child, Computer and Interaction, ISCA, Grenoble, France (Interspeech, ICSLP, Workshop on Child, Computer and Interaction), pp. 18.

G. Hinton , L. Deng , D. Yu , G. E. Dahl , A. Mohamed , N. Jaitly , A. Senior , V. Vanhoucke , P. Nguyen , T. N. Sainath , and B. Kingsbury , 2012. Deep neural networks for acoustic modeling in speech recognition. IEEE Signal Processing Magazine 29 (6): 8297.

G. Hinton , S. Osindero , and Y.-W. Teh , 2006. A fast learning algorithm for deep belief nets. Neural Computation 18 (7): 1527–54.

J. E. Huber , E. T. Stathopoulos , G. M. Curione , T. A. Ash , and K. Johnson , 1999. Formants of children women and men: the effect of vocal intensity variation. Journal of Acoustical Society of America 106 (3): 1532–42.

N. Kumar , and A. G. Andreou , 1998. Heteroscedastic discriminant analysis and reduced rank hmms for improved speech recognition. Speech Communication 26 (4): 283–97.

C.-H. Lee , and J.-L. Gauvain 1993. Speaker adaptation based on map estimation of hmm parameters. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), vol. 2, 558–61.

S. Lee , A. Potamianos , and S. Narayanan , 1999. Acoustic of children's speech: developmental changes of temporal and spectral parameters. Journal of Acoustical Society of America 105 (3): 14551468.

A. Mohamed , G. E. Dahl , and G. Hinton , 2012. Acoustic modeling using deep belief networks. IEEE Transactions on Audio, Speech and Language Processing 20 (1): 1422.

A. Potamianos , and S. Narayanan , 2003. Robust recognition of children's speech. IEEE Transasctions on Speech and Audio Processing 11 (6): 603–15.

S. Steidl , G. Stemmer , C. Hacker , E. Nöth , and H. Niemann , 2003. Improving children's speech recognition by HMM interpolation with an adults’ speech recognizer. In Proceedings of Pattern Recognition, 25th DAGM Symposium, Springer, Berlin, Germany (Pattern Recognition - 25th DAGM Symposium), pp. 600–7.

K. Veselỳ , L. Burget , and F. Grézl 2010. Parallel training of neural networks for speech recognition. In Text, Speech and Dialogue, Springer, Berlin, Germany, pp. 439–46. Springer.

M. Wöllmer , B. Schuller , A. Batliner , S. Steidl , and D. Seppi 2011. Tandem decoding of children's speech for keyword detection in a child-robot interaction scenario. ACM Transasctions Speech Language Processing 7 (4): 12:1–12:22.

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

Natural Language Engineering
  • ISSN: 1351-3249
  • EISSN: 1469-8110
  • URL: /core/journals/natural-language-engineering
Please enter your name
Please enter a valid email address
Who would you like to send this to? *
×

Metrics

Full text views

Total number of HTML views: 29
Total number of PDF views: 65 *
Loading metrics...

Abstract views

Total abstract views: 148 *
Loading metrics...

* Views captured on Cambridge Core between September 2016 - 29th March 2017. This data will be updated every 24 hours.