Skip to main content
×
Home

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 no-reply@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
References
Hide All
Abdel-Hamid O., and Jiang H., 2013a. Fast speaker adaptation of hybrid NN/HMM model for speech recognition based on discriminative learning of speaker code. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 7942–6.
Abdel-Hamid O., and Jiang H., 2013b. Rapid and effective speaker adaptation of convolutional neural network based models for speech recognition. In Proceedings of INTERSPEECH, ISCA, Grenoble, France (Interspeech, ICSLP, Workshop on Child, Computer and Interaction), pp. 1248–52.
Angelini B., Brugnara F., Falavigna D., Giuliani D., Gretter R., and Omologo M., 1994. Speaker independent continuous speech recognition using an acoustic-phonetic Italian corpus. In Proceedings of ICSLP, ISCA, Grenoble, France (Interspeech, ICSLP, Workshop on Child, Computer and Interaction), pp. 1391–4.
Bell P., Swietojanski P., and Renals S., 2013. Multi-level adaptive networks in tandem and hybrid ASR systems. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 6975–9.
Bengio Y., Courville A., and Vincent P., 2013. Representation learning: a review and new perspectives. IEEE Transactions on Pattern Analysis and Machine Intelligence 35 (8): 1798–828.
Bengio Y., Lamblin P., Popovici D., and Larochelle H. 2007. Greedy layer-wise training of deep networks. In Proceedings of NIPS, NIPS, La Jolla (CA), United States, vol. 19, 153–60.
Bourlard H., and Morgan N. 1994. Connectionist Speech Recognition: A Hybrid Approach, Springer, Berlin, Germany, vol. 247, Springer.
Burget L., Schwarz P., Agarwal M., Akyazi P., Feng K., Ghoshal A., Glembek O., Goel N., Karafiat M., Povey D., Rastrow A., Rose R. C., and Thomas S., 2010. Multilingual acoustic modeling for speech recognition based on subspace Gaussian mixture models. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 4334–7.
Claes T., Dologlou I., ten Bosch L., and Van Compernolle D. 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.
Dahl G. E., Yu D., Deng L., and Acero A., 2012. Context-dependent pre-trained deep neural networks for large-vocabulary speech recognition. IEEE Transactions on Audio, Speech and Language Processing 20 (1): 3042.
Das S., Nix D., and Picheny M. 1998. Improvements in Children's speech recognition performance. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN).
Dehak N., Kenny P., Dehak R., Dumouchel P., and Ouellet P., 2011. Front-end factor analysis for speaker verification. IEEE Transactions on Audio, Speech and Language Processing 19 (4): 788–98.
Eide E., and Gish H., 1996. A parametric approach to vocal tract lenght normalization. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 346–9.
Erhan D., Bengio Y., Courville A., Manzagol P.-A., Vincent P., and Bengio S. 2010. Why does unsupervised pre-training help deep learning? The Journal of Machine Learning Research 11 (2): 625–60.
Fitch W. T., and Giedd J., 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.
Gales M. J. F., 1998. Maximum likelihood linear transformations for HMM-based speech recognition. Computer Speech and Language 12 (2): 7598.
Gerosa M., Giuliani D., and F. Brugnara F., 2007. Acoustic variability and automatic recognition of children's speech. Speech Communication 49 (10–11): 847–60.
Gerosa M., Giuliani D., and Brugnara F., 2009a. Towards age-independent acoustic modeling. Speech Communication 51 (6): 499509.
Gerosa M., Giuliani D., Narayanan S., and Potamianos A., 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.
Gillick L., and Cox S. 1989. Some statistical issues in the comparison of speech recognition algorithms. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), 1: 532–5.
Giuliani D., and Gerosa M. 2003. Investigating recognition of children speech. In Proceedings of ICASSP IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), 2: 137–40.
Hagen A., Pellom B., and Cole R. 2003. Children's speech recognition with application to interactive books and tutors. In Proceedings of ASRU. IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN).
Hinton G., Deng L., Yu D., Dahl G. E., Mohamed A., Jaitly N., Senior A., Vanhoucke V., Nguyen P., Sainath T. N., and Kingsbury B., 2012. Deep neural networks for acoustic modeling in speech recognition. IEEE Signal Processing Magazine 29 (6): 8297.
Hinton G., Osindero S., and Teh Y.-W., 2006. A fast learning algorithm for deep belief nets. Neural Computation 18 (7): 1527–54.
Huber J. E., Stathopoulos E. T., Curione G. M., Ash T. A., and Johnson K., 1999. Formants of children women and men: the effect of vocal intensity variation. Journal of Acoustical Society of America 106 (3): 1532–42.
Imseng D., Motlicek P., Garner P. N., and Bourlard H., 2013. Impact of deep MLP architecture on different acoustic modeling techniques for under-resourced speech recognition. In Proceedings of ASRU, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 332–7.
Kumar N., and Andreou A. G., 1998. Heteroscedastic discriminant analysis and reduced rank hmms for improved speech recognition. Speech Communication 26 (4): 283–97.
Le V.-B., Lamel L., and Gauvain J.-L., 2010. Multi-style ML features for BN transcription. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 4866–9.
Lee C.-H., and Gauvain J.-L. 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.
Lee L., and Rose R. C., 1996. Speaker normalization using efficient frequency warping procedure. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 353–6.
Lee S., Potamianos A., and Narayanan S., 1999. Acoustic of children's speech: developmental changes of temporal and spectral parameters. Journal of Acoustical Society of America 105 (3): 14551468.
Li B., and Sim K. C., 2010. Comparison of discriminative input and output transformations for speaker adaptation in the hybrid NN/HMM systems. In Proceedings of INTERSPEECH, ISCA, Grenoble, France (Interspeech, ICSLP, Workshop on Child, Computer and Interaction), pp. 526–9.
Li Q., and Russell M. 2001. Why is automatic recognition of children's speech difficult? In Proceedings of EUROSPEECH. ISCA, Grenoble, France (Interspeech, ICSLP, Workshop on Child, Computer and Interaction).
Liao H., 2013. Speaker adaptation of context dependent deep neural networks. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 7947–51.
Maragakis M. G., and A. Potamianos A., 2008. Region-based vocal tract length normalization for ASR. In Proceedings of INTERSPEECH, ISCA, Grenoble, France (Interspeech, ICSLP, Workshop on Child, Computer and Interaction), pp. 1365–8.
Metallinou A., and Cheng J., 2014. Using deep neural networks to improve proficiency assessment for children english language learners. In Proceedings of INTERSPEECH, ISCA, Grenoble, France (Interspeech, ICSLP, Workshop on Child, Computer and Interaction), pp. 1468–72.
Miguel A., Lieida E., Rose R., Buera L., and Ortega A., 2005. Augmented state space acoustic decoding for modeling local variability in speech. In Proceedings of INTERSPEECH, ISCA, Grenoble, France (Interspeech, ICSLP, Workshop on Child, Computer and Interaction), pp. 3009–12.
Mohamed A., Dahl G. E., and Hinton G., 2012. Acoustic modeling using deep belief networks. IEEE Transactions on Audio, Speech and Language Processing 20 (1): 1422.
Nisimura R., Lee A., Saruwatari H., and Shikano K. 2004. Public speech-oriented guidance system with adult and child discrimination capability. In Proceedings of ICASSP. IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN).
Pinto J., Magimai-Doss N., and Bourlard H., 2009. MLP based hierarchical system for task adaptation in ASR. In Proceedings of ASRU, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 365–70.
Potamianos A., and Narayanan S., 2003. Robust recognition of children's speech. IEEE Transasctions on Speech and Audio Processing 11 (6): 603–15.
Saon G., Soltau H., Nahamoo D., and Picheny M., 2013. Speaker adaptation of neural network acoustic models using I-vectors. In Proceedings of ASRU, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 55–9.
Seide F., Li G., Chen X., and Yu D., 2011. Feature engineering in context-dependent deep neural networks for conversational speech transcription. In Proceedings of ASRU, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 24–9.
Seltzer M. L., Yu D., and Wang Y. 2013. An investigation of deep neural networks for noise robust speech recognition. In Proceedings of ICASSP. IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN).
Senior A., and Lopez-Moreno I. 2014. Improving DNN speaker independence with I-vector inputs. In Proceedings of ICASSP. IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN).
Serizel R., and Giuliani D. 2014a. Deep neural network adaptation for children's and adults’ speech recognition. In Proceedings of CLIC-It. Pisa University Press, Pisa, Italy (CLIC-It).
Serizel R., and Giuliani D. 2014b. Vocal tract length normalisation approaches to DNN-based children's and adults’ speech recognition. In Proceedings of SLT. IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN).
Sivadas S., and Hermansky H. 2004. On use of task independent training data in tandem feature extraction. In Proceedings of ICASSP vol. 1, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 541–4.
Steidl S., Stemmer G., Hacker C., Nöth E., and Niemann H., 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.
Stolcke A., Grézl F., Hwang M.-Y., Lei X., Morgan N., and Vergyri D., 2006. Cross-domain and cross-language portability of acoustic features estimated by multilayer perceptrons. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), vol. 1, pp. 321–34.
Swietojanski P., Ghoshal A., and Renals S., 2012. Unsupervised cross-lingual knowledge transfer in DNN-based LVCSR. In Proceedings of SLT, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 246–51.
Thomas S., Seltzer M. L., Church K., and Hermansky H., 2013. Deep neural network features and semi-supervised training for low resource speech recognition. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 6704–8.
Veselỳ K., Burget L., and Grézl F. 2010. Parallel training of neural networks for speech recognition. In Text, Speech and Dialogue, Springer, Berlin, Germany, pp. 439–46. Springer.
Wegmann S., McAllaster D., Orloff J., and Peskin B. 1996. Speaker normalisation on conversational telephone speech. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), vol. 1, 339–41.
Welling L., Kanthak S., and Ney H. 1999. Improved methods for vocal tract normalization. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), vol. 2, 761–4.
Wilpon J. G., and Jacobsen C. N., 1996. A study of speech recognition for children and elderly. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), pp. 349–52.
Wöllmer M., Schuller B., Batliner A., Steidl S., and Seppi D. 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.
Woodland P. C., Odell J. J., Valtchev V., and Young S. J., 1994. Large vocabulary continuous speech recognition using HTK. In Proceedings of ICASSP, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), vol. 2, pp. 125–8.
Yochai K., and Morgan N., 1992. GDNN: a gender-dependent neural network for continuous speech recognition. In Proceedings of Iternational Joint Conference on Neural Networks, IEEE, New York (NY), United States (ICASSP, ASRU, SLT, IJCNN), vol. 2, pp. 332–7.
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: 39
Total number of PDF views: 208 *
Loading metrics...

Abstract views

Total abstract views: 360 *
Loading metrics...

* Views captured on Cambridge Core between September 2016 - 23rd November 2017. This data will be updated every 24 hours.