Your browser does not support JavaScript!
http://iet.metastore.ingenta.com
1887

access icon openaccess Speech reconstruction using a deep partially supervised neural network

Statistical speech reconstruction for larynx-related dysphonia has achieved good performance using Gaussian mixture models and, more recently, restricted Boltzmann machine arrays; however, deep neural network (DNN)-based systems have been hampered by the limited amount of training data available from individual voice-loss patients. The authors propose a novel DNN structure that allows a partially supervised training approach on spectral features from smaller data sets, yielding very good results compared with the current state-of-the-art.

References

    1. 1)
      • 8. Kawahara, H., Morise, M., Takahashi, T., et al: ‘TANDEM-STRAIGHT: a temporally stable power spectral representation for periodic signals and applications to interference-free spectrum, f0, and aperiodicity estimation’. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, 2008 ICASSP 2008, 2008, pp. 39333936.
    2. 2)
      • 4. McLoughlin, I.V., Li, J., Song, Y.: ‘Reconstruction of continuous voiced speech from whispers’. Proc. Interspeech, August 2013, pp. 10221026.
    3. 3)
      • 15. McLoughlin, I.V.: ‘Speech and audio processing: a MATLAB-based approach’ (Cambridge University Press, 2016).
    4. 4)
    5. 5)
    6. 6)
    7. 7)
    8. 8)
    9. 9)
    10. 10)
    11. 11)
      • 1. Mcloughlin, I.V., Sharifzadeh, H.R., Tan, S.L., et al: ‘Reconstruction of phonated speech from whispers using formant-derived plausible pitch modulation’, ACM Trans. Accessible Comput. (TACCESS), 2015, 6, (4), p. 12.
    12. 12)
      • 7. Tajiri, Y., Tanaka, K., Toda, T., et al: ‘Non-audible murmur enhancement based on statistical conversion using air- and body-conductive microphones in noisy environments’. 16th Annual Conf. of the Int. Speech Communication Association, 2015.
    13. 13)
    14. 14)
    15. 15)
http://iet.metastore.ingenta.com/content/journals/10.1049/htl.2016.0103
Loading

Related content

content/journals/10.1049/htl.2016.0103
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading
This is a required field
Please enter a valid email address