Abstract
Robust and accurate predicting of articulatory movements has various important applications, such as human-machine interaction. Various approaches have been proposed to solve the acoustic-articulatory mapping problem. However, their precision is not high enough with only acoustic features available. Recently, deep neural network (DNN) has brought tremendous success in many fields. To increase the accuracy, on the one hand, we propose a new network architecture called bottleneck squeeze-and-excitation recurrent convolutional neural network (BSERCNN) for articulatory movement prediction. On the one hand, by introducing the squeeze-and-excitation (SE) module, our BSERCNN can model the interdependencies and relationships between channels and that makes our model more efficiency. On the other hand, phoneme-level text features and acoustic features are integrated together as inputs to BSERCNN for better performance. Experiments show that BSERCNN achieves the state-of-the-art root-mean-squared error (RMSE) 0.563 mm and the correlation coefficient 0.954 with both text and audio inputs.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The demo can be downloaded from https://pan.baidu.com/s/1q0YLvdqq8WVU5OOLj0ru1w.
References
Yu, J., Wang, Z.-F.: A video, text, and speech-driven realistic 3-D virtual head for human-machine interface. IEEE Trans. Cybern. 45(5), 991–1002 (2015)
Zhao, G., Barnard, M., Pietikainen, M.: Lipreading with local spatiotemporal descriptors. IEEE Trans. Multimedia 11(7), 1254–1265 (2009)
Fanelli, G., Gall, J., Romsdorfer, H., Weise, T., Van Gool, L.: A 3-D audio-visual corpus of affective communication. IEEE Trans. Multimedia 12(6), 591–598 (2010)
Mitra, V.: Articulatory information for robust speech recognition, Ph.D. dissertation (2010)
Toda, T., Black, A.W., Tokuda, K.: Statistical mapping between articulatory movements and acoustic spectrum using a gaussian mixture model. Speech Commun. 50(3), 215–227 (2008)
Zhang, L., Renals, S.: Acoustic-articulatory modeling with the trajectory HMM. IEEE Signal Process. Lett. 15, 245–248 (2008)
Deng, L., Hinton, G., Kingsbury, B.: New types of deep neural network learning for speech recognition and related applications: an overview. In: IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 8599–8603 (2013)
Qian, Y., Fan, Y., Hu, W., Soong, F.K.: On the training aspects of deep neural network (DNN) for parametric TTS synthesis. In: IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 3829–3833 (2014)
Uria, B., Murray, I., Renals, S., Richmond, K.: Deep architectures for articulatory inversion. In: Thirteenth Annual Conference of the International Speech Communication Association (2012)
Uria, B., Renals, S., Richmond, K.: A deep neural network for acoustic-articulatory speech inversion. In: NIPS 2011 Workshop on Deep Learning and Unsupervised Feature Learning (2011)
Zhu, P., Xie, L., Chen, Y.: Articulatory movement prediction using deep bidirectional long short-term memory based recurrent neural networks and word/phone embeddings. In: INTERSPEECH, pp. 2192–2196 (2015)
Wei, Z., Wu, Z., Xie, L.: Predicting articulatory movement from text using deep architecture with stacked bottleneck features. In: 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA), pp. 1–6. IEEE (2016)
Ling, Z.H., Richmond, K., Yamagishi, J.: An analysis of HMM-based prediction of articulatory movements. Speech Commun. 52(10), 834–846 (2010)
Abdel-Hamid, O., Mohamed, A.-R., Jiang, H., Deng, L., Penn, G., Yu, D.: Convolutional neural networks for speech recognition. IEEE/ACM Trans. Audio, Speech, Lang. Process. 22(10), 1533–1545 (2014)
Hu, J., Shen, L., Sun, G.: Squeeze-and-excitation networks, arXiv preprint arXiv:1709.01507, vol. 7 (2017)
Yu, D., Seltzer, M.L.: Improved bottleneck features using pretrained deep neural networks. In: Twelfth Annual Conference of the International Speech Communication Association (2011)
Cheng, X., Li, X., Tai, Y., Yang, J.: SESR: Single image super resolution with recursive squeeze and excitation networks, arXiv preprint arXiv:1801.10319 (2018)
Schönle, P.W., Gräbe, K., Wenig, P., Höhne, J., Schrader, J., Conrad, B.: Electromagnetic articulography: use of alternating magnetic fields for trackingmovements of multiple points inside and outside the vocal tract. Brain Lang. 31(1), 26–35 (1987)
Wu, Z., Watts, O., King, S.: Merlin: an open source neural network speech synthesis system. Proc. SSW, Sunnyvale, USA (2016)
Jia, Y., Shelhamer, E., Donahue, J., et al.: Caffe: Convolutional architecture for fast feature embedding. In: Proceedings of the 22nd ACM International Conference on Multimedia, pp. 675–678. ACM (2014)
Ling, Z.-H., Richmond, K., Yamagishi, J.: HMM-based text-to-articulatory-movement prediction and analysis of critical articulators. In: Proc. Interspeech, pp. 2194–2197, Sep. 2010
Richmond, K.: Preliminary inversion mapping results with a new EMA corpus (2009)
Liu, P., Yu, Q., Wu, Z., Kang, S., Meng, H., Cai, L.: A deep recurrent approach for acoustic-to-articulatory inversion. In: 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 4450–4454. IEEE (2015)
Yu, J., Li, A., Hu, F., et al.: Data-driven 3D visual pronunciation of Chinese IPA for language learning. In: 2013 International Conference on Oriental COCOSDA Held Jointly with 2013 Conference on Asian Spoken Language Research and Evaluation (O-COCOSDA/CASLRE), pp. 1–6. IEEE (2013)
Marcos, S., Gómez-García-Bermejo, J., Zalama, E.: A realistic, virtual head for human-computer interaction. Interact. Comput. 22(3), 176–192 (2010)
Acknowledgement
This work is supported by the National Natural Science Foundation of China (U1736123, 61572450), Anhui Provincial Natural Science Foundation (1708085QF138), the Fundamental Research Funds for the Central Universities (WK2350000002).
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Yu, L., Yu, J., Ling, Q. (2019). Deep Neural Network Based 3D Articulatory Movement Prediction Using Both Text and Audio Inputs. In: Kompatsiaris, I., Huet, B., Mezaris, V., Gurrin, C., Cheng, WH., Vrochidis, S. (eds) MultiMedia Modeling. MMM 2019. Lecture Notes in Computer Science(), vol 11295. Springer, Cham. https://doi.org/10.1007/978-3-030-05710-7_6
Download citation
DOI: https://doi.org/10.1007/978-3-030-05710-7_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-05709-1
Online ISBN: 978-3-030-05710-7
eBook Packages: Computer ScienceComputer Science (R0)