A Novel Recurrent Generalized Congruence Neural Network Measure for Objective Speech Quality Evaluation

Abstract:

A new system model for objective speech quality evaluation based on the improved recurrent generalized congruence neural network (RGCNN/OSQE) is proposed. The performance of the RGCNN model is compared with the most commonly used RBFNN (radial basis function neural network) model in objective speech quality evaluation. Comparison results show that the RGCNN model has higher correlation coefficient, less deviation, and saves about half training time, i.e., the RGCNN model has obvious advantages over the RBFNN model. Therefore, the novel RGCNN model for objective speech quality evaluation is feasible and effective.

• Periodical: Advanced Materials Research (Volumes 433-440)
• Pages: 2282-2287
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.433-440.2282
• Online since: January 2012
• Authors: Tian Yun Yan
• Keywords: MOS, RBF, Recurrent Neural Network, Speech Quality

References

[1] Guo Chen, Xiulin Hu, Yunyu Zhang, Yaoting Zhu, Research advance on objective measures of speech quality, Acta Electronica Sinica, 2001, vol. 29, no. 4, pp.548-552. (In Chinese).

[2] D. Sliskovic, E. K. Nyarko, N. Peric, Estimation of difficult-to-measure process variables using neural networks - a comparison of simple MLP and RBF neural network properties, Proceedings of the 12th IEEE Mediterranean Electrotechnical Conference, 2004, vol. 1, pp.387-390.

DOI: 10.1109/melcon.2004.1346888

[3] E.H. Lim, K.P. Seng, K.M. Tse, RBF neural network mouth tracking for audio-visual speech recognition system, IEEE Region 10 Conference on TENCON, 2004, Vol. A, pp.84-87.

DOI: 10.1109/tencon.2004.1414362

[4] Qiang Fu, Bin Tian, Zhiyi Zhang, Kechu Yi, Research on speech distortion measure based on neural network, Acta Acustica, 2001, vol. 26, no. 2, pp.180-184. (In Chinese).

[5] Qiang Fu, Kechu Yi, Mingui Sun, Speech quality objective assessment using neural network, Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing, 2000, vol. 3, pp.1511-1514.

DOI: 10.1109/icassp.2000.861932

[6] Tianyun Yan, Xia Yun, Fan Jin, Qingjun Zhu, RBF neural networks and their application to output-based objective speech quality assessment, Acta Electronica Sinica, 2004, vol. 32, no. 8, pp.1282-1285. (In Chinese).

[7] Tianyun Yan, Hefei Ling, Shurong Zou, A novel recurrent generalized congruence neural network for dynamical system identification, Proceeding of the 3rd International Conference on Natural Computation and the 4th International Conference on Fuzzy Systems and Knowledge Discovery, China, 2007, vol. 1, pp.39-43.

DOI: 10.1109/icnc.2007.122