The Journal of the Acoustical Society of Korea (한국음향학회지)

The Acoustical Society of Korea (한국음향학회)
권호 표지
DOI QR코드

DOI QR Code

Transformer-based transfer learning and multi-task learning for improving the performance of speech emotion recognition

음성감정인식 성능 향상을 위한 트랜스포머 기반 전이학습 및 다중작업학습

  • 박순찬 (부산대학교 전자공학과) ;
  • 김형순 (부산대학교 전자공학과)
  • Received : 2021.07.16
  • Accepted : 2021.08.25
  • Published : 2021.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

It is hard to prepare sufficient training data for speech emotion recognition due to the difficulty of emotion labeling. In this paper, we apply transfer learning with large-scale training data for speech recognition on a transformer-based model to improve the performance of speech emotion recognition. In addition, we propose a method to utilize context information without decoding by multi-task learning with speech recognition. According to the speech emotion recognition experiments using the IEMOCAP dataset, our model achieves a weighted accuracy of 70.6 % and an unweighted accuracy of 71.6 %, which shows that the proposed method is effective in improving the performance of speech emotion recognition.

음성감정인식을 위한 훈련 데이터는 감정 레이블링의 어려움으로 인해 충분히 확보하기 어렵다. 본 논문에서는 음성감정인식의 성능 개선을 위해 트랜스포머 기반 모델에 대규모 음성인식용 훈련 데이터를 통한 전이학습을 적용한다. 또한 음성인식과의 다중작업학습을 통해 별도의 디코딩 없이 문맥 정보를 활용하는 방법을 제안한다. IEMOCAP 데이터 셋을 이용한 음성감정인식 실험을 통해, 가중정확도 70.6 % 및 비가중정확도 71.6 %를 달성하여, 제안된 방법이 음성감정인식 성능 향상에 효과가 있음을 보여준다.

Keywords

Acknowledgement

이 논문은 2019년교육부와 한국연구재단의 지원을 받아 수행된 연구임(NRF-2019S1A5A2A03045884)

References

  1. H. Hu, M. Xu, and W. Wu, "GMM supervector based SVM with spectral features for speech emotion recognition," Proc. ICASSP. 413-416 (2007).
  2. A. Stuhlsatz, C. Meyer, F. Eyben, T. Zielke, G. Meier, and B. Schuller, "Deep neural networks for acoustic emotion recognition: Raising the benchmarks," Proc. ICASSP. 5688-5691 (2011).
  3. G. Trigeorgis, F. Ringeval, R. Brueckner, E. Marchi, M. A. Nicolaou, B. Schuller, and S. Zafeiriou, "Adieu features? End-to-end speech emotion recognition using a deep convolutional recurrent network," Proc. ICASSP. 5200-5204 (2016).
  4. S. Mirsamadi, E. Barsoum, and C. Zhang, "Automatic speech emotion recognition using recurrent neural networks with local attention," Proc. ICASSP. 2227-2231 (2017).
  5. J. Kim, G. Englebienne, K. P. Truong, and V. Eversu, "Towards speech emotion recognition "in the Wild" using aggregated corpora and deep multi-task learning," Proc. Interspeech, 1113-1117 (2017).
  6. S. Yoon, S. Byun, and K. Jung, "Multimodal speech emotion recognition using audio and text," Proc. SLT. 112-118 (2018).
  7. Z. Lu, L. Cao, Y. Zhang, C. Chiu, and J. Fan, "Speech sentiment analysis via pre-trained features from end-to-end ASR models," Proc. ICASSP. 7149-7153 (2020).
  8. A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, and L.Kaiser, "Attention is all you need," Proc. NIPS. 6000-6010 (2017).
  9. J. Devlin, M. Chang, K. Lee, and K. Toutanova, "BERT: Pre-training of deep bidirectional transformers for language understanding," Proc. NAACL-HLT. 4171-4186 (2019).
  10. A. Baevski, Y. Zhou, A. Mohamed, and M. Auli, "Wav2vec 2.0: A framework for self-supervised learning of speech representations," Proc. NeurIPS. 12449-12460 (2020).
  11. C. Busso, M. Bulut, C.-C. Lee, A. Kazemzadeh, E. Mower, S.l Kim, J. N. Chang, S. Lee, and S. S. Narayanan, "IEMOCAP: interactive emotional dyadic motion capture database," Language Resources and Evaluation, 42, 335-359 (2008). https://doi.org/10.1007/s10579-008-9076-6
  12. V. Panayotov, G. Chen, D. Povey, and S. Khudanpur, "Librispeech: An ASR corpus based on public domain audio books," Proc. ICASSP. 5206-5210 (2015).
  13. W. Chan, N. Jaitly, Q. Le, and O. Vinyals, "Listen, attend and spell: A neural network for large vocabulary conversational speech recognition," Proc. ICASSP. 4960-4964 (2016).
  14. A. Graves, A. Mohamed, and G. Hinton, "Speech recognition with deep recurrent neural networks," Proc. ICASSP. 6645-6649 (2013).
  15. A. Graves, S. Fernandez, F. Gomez, and J. Schmidhuber, "Connectionist temporal classification: Labelling unsegmented sequence data with recurrent neural networks," Proc. ICML. 369-376 (2006).
  16. S. Watanabe, T. Hori, S. Kim, J. R. Hershey, and T. Hayashi, "Hybrid CTC/attention architecture for end-to-end speech recognition," IEEE JSTSP. 11, 1240-1253 (2017).
  17. T. Kudo and J. Richardson, "SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing," Proc. EMNLP 66-71 (2018).
  18. A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, T. Killeen, Z. Lin, N. Gimelshein, L. Antiga, A. Desmaison, A. Kopf, E. Yang, Z. DeVito, M. Raison, A. Tejani, S. Chilamkurthy, B. Steiner, L. Fang, J. Bai, and S. Chintala, "PyTorch: An imperative style, high-performance deep learning library," Proc. NeurIPS. 8024-8035 (2019).