A Step Towards Preserving Speakers' Identity While Detecting Depression Via Speaker Disentanglement

Federated learning and deep learning framework for MRI image and speech signal-based multi-modal depression detection

Adolescence is a significant period for developing skills and knowledge and learning about managing relationships and emotions by gathering attributes for maturity. Recently, Depression arises as a common mental health issue in adolescents and this affects the daily life of the person. This leads to educational and social impairments and this acts as a major risk for suicide. As a result, the identification and treatment for this disorder are essential. By applying Deep learning (DL) algorithms to medical data, the mental condition of a person can be predicted. However, the traditional deep learning models face the challenge in processing the huge sized data. Hence, FL has emerged as an efficient solution for addressing the data size issue of DL. Here, Depression detection in adolescents is carried out by considering the FL framework, which comprises two modules, namely the local module and the Global module. The detection process is done in the local module using the proposed Exponential African Pelican Optimization based Deep Convolutional Neural Network (ExpAPO-DCNN), whereas the Global module produces the aggregated output of the local module. In this research, FL utilizes the DL model in producing the output, where the DL model considered two modalities of inputs, such as speech signal and Magnetic Resonance Imaging (MRI) image. The processing steps used for this research are pre-processing, feature extraction and detection. For MRI and speech signals, all the above processes are carried out individually. Finally, both the outputs are fused utilizing the overlap coefficient. The ExpAPO-DCNN obtained accuracy, Loss, Root mean Squared error (RMSE), Mean Squared error (MSE), True Negative rate (TNR), and True Positive rate (TPR) of 98.00 %, 0.023, 0.058, 0.240, 97.90 %, and 96.30 %, respectively.

Diagnostic accuracy of deep learning using speech samples in depression: a systematic review and meta-analysis

OBJECTIVE

This study aims to conduct a systematic review and meta-analysis of the diagnostic accuracy of deep learning (DL) using speech samples in depression.

MATERIALS AND METHODS

This review included studies reporting diagnostic results of DL algorithms in depression using speech data, published from inception to January 31, 2024, on PubMed, Medline, Embase, PsycINFO, Scopus, IEEE, and Web of Science databases. Pooled accuracy, sensitivity, and specificity were obtained by random-effect models. The diagnostic Precision Study Quality Assessment Tool (QUADAS-2) was used to assess the risk of bias.

RESULTS

A total of 25 studies met the inclusion criteria and 8 of them were used in the meta-analysis. The pooled estimates of accuracy, specificity, and sensitivity for depression detection models were 0.87 (95% CI, 0.81-0.93), 0.85 (95% CI, 0.78-0.91), and 0.82 (95% CI, 0.71-0.94), respectively. When stratified by model structure, the highest pooled diagnostic accuracy was 0.89 (95% CI, 0.81-0.97) in the handcrafted group.

DISCUSSION

To our knowledge, our study is the first meta-analysis on the diagnostic performance of DL for depression detection from speech samples. All studies included in the meta-analysis used convolutional neural network (CNN) models, posing problems in deciphering the performance of other DL algorithms. The handcrafted model performed better than the end-to-end model in speech depression detection.

CONCLUSIONS

The application of DL in speech provided a useful tool for depression detection. CNN models with handcrafted acoustic features could help to improve the diagnostic performance.

PROTOCOL REGISTRATION

The study protocol was registered on PROSPERO (CRD42023423603).

Improving speech depression detection using transfer learning with wav2vec 2.0 in low-resource environments

Depression, a pervasive global mental disorder, profoundly impacts daily lives. Despite numerous deep learning studies focused on depression detection through speech analysis, the shortage of annotated bulk samples hampers the development of effective models. In response to this challenge, our research introduces a transfer learning approach for detecting depression in speech, aiming to overcome constraints imposed by limited resources. In the context of feature representation, we obtain depression-related features by fine-tuning wav2vec 2.0. By integrating 1D-CNN and attention pooling structures, we generate advanced features at the segment level, thereby enhancing the model's capability to capture temporal relationships within audio frames. In the realm of prediction results, we integrate LSTM and self-attention mechanisms. This incorporation assigns greater weights to segments associated with depression, thereby augmenting the model's discernment of depression-related information. The experimental results indicate that our model has achieved impressive F1 scores, reaching 79% on the DAIC-WOZ dataset and 90.53% on the CMDC dataset. It outperforms recent baseline models in the field of speech-based depression detection. This provides a promising solution for effective depression detection in low-resource environments.

Non-uniform Speaker Disentanglement For Depression Detection From Raw Speech Signals

While speech-based depression detection methods that use speaker-identity features, such as speaker embeddings, are popular, they often compromise patient privacy. To address this issue, we propose a speaker disentanglement method that utilizes a non-uniform mechanism of adversarial SID loss maximization. This is achieved by varying the adversarial weight between different layers of a model during training. We find that a greater adversarial weight for the initial layers leads to performance improvement. Our approach using the ECAPA-TDNN model achieves an F1-score of 0.7349 (a 3.7% improvement over audio-only SOTA) on the DAIC-WoZ dataset, while simultaneously reducing the speaker-identification accuracy by 50%. Our findings suggest that identifying depression through speech signals can be accomplished without placing undue reliance on a speaker's identity, paving the way for privacy-preserving approaches of depression detection.