Search papers, labs, and topics across Lattice.
This study introduces CLeaD, a supervised contrastive alignment framework that effectively maps WavLM embeddings from English and Mandarin into a shared clinical space for depression detection. By addressing identity leakage in prior methodologies, the approach demonstrates a modest improvement in F1 score (0.640 vs. 0.622) in a cross-lingual context, while also revealing that model scaling negatively impacts cross-lingual performance. Additionally, the research highlights the inflated F1 scores in previous studies due to speaker identity leakage, providing critical insights into the reliability of cross-lingual depression detection metrics.
Identity leakage previously inflated Mandarin depression detection scores to 0.954, but CLeaD reveals the true performance is significantly lower, exposing critical flaws in existing methodologies.
Significant disparities exist in the diagnosis and clinical presentation of depression across different linguistic populations. Speech-based depression detection performs well monolingually, but cross-lingual generalization remains an open challenge. A key reason is that prior work uses segment-level random splits without speaker grouping, leading to identity leakage that inflates reported metrics. We propose CLeaD, a supervised contrastive alignment framework that maps WavLM embeddings from English and Mandarin into a shared clinical space, without parallel data or target-language fine-tuning. Evaluating 52 Mandarin speakers, contrastive alignment modestly outperforms the baseline (F1: 0.640 vs. 0.622) under leave-one-speaker-out evaluation. It also improves depressed-class recall at intermediate layers (7-8), though the small test set limits generalizability. Two findings remain robust: model scaling degrades cross-lingual performance while improving monolingual English, and speaker identity leakage artificially inflated previously reported Mandarin F1 scores to 0.954, an artifact we reproduce and quantify.
