基於隱藏式條件隨機域聲學模型之強健式華英混雜語音 辨認演算法

Mixed-Lingual Acoustic Modeling of Hidden Conditional Random Field for Robust Speech Recognition

黃詩涵(Shih-Han Huang)

指導教授 ： 洪維廷

元智大學/電機通訊學院/通訊工程學系/碩士(2011年)

摘要

本論文提出以強健式演算法(論文中簡稱REST)訓練隱藏式條件隨機域(Hidden　Conditional Random Fields，簡稱HCRF)華語/英語聲學模型，嘗試解決(1)混
雜語音語音辨認之抗雜訊問題和(2)混雜語音語音辨認之跨語系辨認錯誤問題。REST演算法可以提高HCRF模型對雜訊環境的辨識效能，接著透過鑑別式法則訓練HCRF模型，提升HCRF模型對語音模型的鑑別能力，並且大幅降低跨語言語音辨認之錯誤。根據一連串之實驗證明，基於HCRF語音模型之錯誤率平均值比傳統HMM降低約16.41%(Rover_2雜訊)，並且跨語言語音辨認之錯誤大幅降低。

關鍵字

隱藏式條件隨機域聲學模型 ； 語音辨識 ； 強健式訓練

並列摘要

This thesis presents the robust training techniques for hidden conditional random fiels (HCRF)-based acoustic modeling of Mandarin/English mixed-lingual speech recognition. Two issues were dealt with: (1) mixed-lingual speech recognition against with noise effects and (2) cross-lingual errors in mixed-lingual speech recognition. We solved first issue with the REST algorithm and reduce the errors in second issue with a discriminative training algorithm combined by the REST algorithm(D-REST). The experimental results indicate that 16.4% averaged error rate reduction by the HCRF-based framework is achieved under ROVER_2 noise environment compared with the result by the traditional HMM approach. In additional, the cross-lingual error is improved significantly with the HCRF-framework in mixed-lingual speech recognition.

並列關鍵字

HMM ； HCRF ； Robust Training Algorithm

參考文獻

[1] L. Rabiner, “A tutorial on hidden markov models and selected applications in　speech recognition,” Proceedings of the IEEE, vol. 77, pp. 257–286, 1989.

[2] A. Varga and R. Moore, “Hidden markov model　decomposition of speech and　noise,” in Proceedings International Conference on Acoustics, Speech, and Signal

[5] Y.-H. Sung, C. Boulis, C. Manning, and D. Jurafsky, “Regularization, adaptation,　and non-independent features improve hidden conditional random fields　for phone　classification,” in Proceedings IEEE Workshop on Automatic Speech　Recognition & Understanding, pp. 347–352, 2007.

[6] B. H. Juang and S. Katagirl, “Discriminative learning for minimum error classification,”　Signal Processing, IEEE Transactions on, vol. 40, pp. 3043–3054,1992.

[8] W.-T. Hong and S.-H. Chen, “A robust training algorithm for adverse speech　recognition,” Speech Communication., vol. 30, pp. 273–293, 2000.