ABSTRACT
Voice-enabled communication is increasingly being used in real-world applications, such as the ones involving conversational bots or "chatbots". Chatbots can spark and sustain user engagement by effectively recognizing their emotions and acting upon them. However, the majority of emotion recognition systems rely on rich spectrotemporal acoustic features. Beyond the emotion-related information, such systems tend to preserve information relevant to the identity of the speaker, therefore raising major privacy concerns from the users. This paper introduces two hybrid architectures for privacy-preserving emotion recognition from speech. These architectures rely on a Siamese neural network, whose input and intermediate layers are transformed using various privacy-performing operations in order to retain emotion-dependent content and suppress information related to the identity of a speaker. The proposed approach is evaluated through emotion classification and speaker identification performance metrics. Results indicate that the proposed framework can achieve up to 67.4% for classifying between happy, sad, frustrated, anger, neutral and other emotions, obtained from the publicly available Interactive Emotional Dyadic Motion Capture (IEMOCAP) dataset. At the same time, the proposed approach reduces speaker identification accuracy to 50%, compared to 81%, the latter being achieved through a feedforward neural network solely trained on the speaker identification task using the same input features.
- R Beale and C Peter. 2008. Affect and emotion in human-computer interaction. Springer. Google ScholarDigital Library
- Carlos Busso, Murtaza Bulut, Chi-Chun Lee, Abe Kazemzadeh, Emily Mower, Samuel Kim, Jeannette N Chang, Sungbok Lee, and Shrikanth S Narayanan. 2008. IEMOCAP: Interactive emotional dyadic motion capture database. Language resources and evaluation 42, 4 (2008), 335.Google Scholar
- Houwei Cao, David G Cooper, Michael K Keutmann, Ruben C Gur, Ani Nenkova, and Ragini Verma. 2014. CREMA-D: Crowd-sourced emotional multimodal actors dataset. IEEE transactions on affective computing 5, 4 (2014), 377--390.Google Scholar
- Farah Chenchah and Zied Lachiri. 2014. Speech emotion recognition in acted and spontaneous context. Procedia Computer Science 39 (2014), 139--145.Google ScholarCross Ref
- Scot Cunningham and Traian Marius Truta. 2008. Protecting privacy in recorded conversations. (mar 2008).Google Scholar
- Florian Eyben, Martin Wöllmer, and Björn Schuller. 2010. Opensmile: the munich versatile and fast open-source audio feature extractor. In Proceedings of the 18th ACM international conference on Multimedia. ACM, 1459--1462. Google ScholarDigital Library
- Asbjørn Følstad and Petter Bae Brandtzæg. 2017. Chatbots and the new world of HCI. interactions 24, 4 (2017), 38--42. Google ScholarDigital Library
- Cornelius Glackin, Gerard Chollet, Nazim Dugan, Nigel Cannings, Julie Wall, Shahzaib Tahir, Indranil Ghosh Ray, and Muttukrishnan Rajarajan. 2017. Privacy preserving encrypted phonetic search of speech data. (mar 2017).Google Scholar
- Mohammad Hadian, Thamer Altuwaiyan, Xiaohui Liang, and Wei Li. 2017. Privacy-preserving voice-based search over mHealth data. (aug 2017).Google Scholar
- Jihun Hamm. 2017. Enhancing utility and privacy with noisy minimax filter. (june 2017).Google Scholar
- Daniel Jurafsky and James H Martin. 2017. Dialog Systems and Chatbots. In Speech and language processing. Chapter 29.Google Scholar
- Nir Kshetri and Jeffrey Voas. 2018. Cyberthreats under the Bed. Computer 51, 5 (2018), 92--95.Google ScholarCross Ref
- Chi-Chun Lee, Emily Mower, Carlos Busso, Sungbok Lee, and Shrikanth Narayanan. 2011. Emotion recognition using a hierarchical binary decision tree approach. Speech Communication 53, 9--10 (2011), 1162--1171. Google ScholarDigital Library
- Kun Liu, H. Kargupta, and J. Ryan. 2005. Random projection-based multiplicative data perturbation for privacy preserving distributed data mining. (dec 2005).Google Scholar
- Steven R Livingstone and Frank A Russo. 2018. The Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS): A dynamic, multimodal set of facial and vocal expressions in North American English. PloS one 13, 5 (2018), e0196391.Google ScholarCross Ref
- Lingjuan Lyu, Xuanli He, Yee Wei Law, and Marimuthu Palaniswami. 2017. Privacy-Preserving Collaborative Deep Learning with Application to Human Activity Recognition. In Proceedings of the 2017 ACM on Conference on Information and Knowledge Management. ACM, 1219--1228. Google ScholarDigital Library
- Lingjuan Lyu, Yee Wei Law, Sarah M. Erfani, Christopher Leckie, and Marimuthu Palaniswami. 2016. An improved scheme for privacy-preserving collaborative anomaly detection. (apr 2016).Google Scholar
- Manas A. Pathak, Bhiksha Raj, Shantanu D. Rane, and Paris Smaragdis. 2013. Privacy preserving speech processing. (feb 2013).Google Scholar
- Yogachandran Rahulamathavan and Muttukrishnan Rajarajan. 2015. Efficient privacy-preserving facial expression classification. (may 2015).Google Scholar
- Srinivasan Ramakrishnan and Ibrahiem MM El Emary. 2013. Speech emotion recognition approaches in human computer interaction. Telecommunication Systems 52, 3 (2013), 1467--1478. Google ScholarDigital Library
- LATANYA SWEENEY. 2002. A model for protecting privacy. (may 2002).Google Scholar
- Dimitrios Ververidis and Constantine Kotropoulos. 2006. Emotional speech recognition: Resources, features, and methods. Speech communication 48, 9 (2006), 1162--1181.Google Scholar
- Wikipedia. {n. d.}. https://en.wikipedia.org/wiki/Secure_Hash_Algorithms.Google Scholar
Recommendations
Speech Emotion Recognition via Contrastive Loss under Siamese Networks
ASMMC-MMAC'18: Proceedings of the Joint Workshop of the 4th Workshop on Affective Social Multimedia Computing and first Multi-Modal Affective Computing of Large-Scale Multimedia DataSpeech emotion recognition is an important aspect of human-computer interaction. Prior work proposes various end-to-end models to improve the classification performance. However, most of them rely on the cross-entropy loss together with softmax as the ...
Speech emotion recognition: two decades in a nutshell, benchmarks, and ongoing trends
Tracing 20 years of progress in making machines hear our emotions based on speech signal properties.
Preserving Privacy in Image-based Emotion Recognition through User Anonymization
ICMI '20: Proceedings of the 2020 International Conference on Multimodal InteractionThe large amount of data captured by ambulatory sensing devices can afford us insights into longitudinal behavioral patterns, which can be linked to emotional, psychological, and cognitive outcomes. Yet, the sensitivity of behavioral data, which ...
Comments