arXiv subjects

Zhitong Zhou

Publications and source records attributed to Zhitong Zhou.

3 recordsLinked to original sources

arXiv2026

Depictions of Depression in Generative AI Video Models: A Preliminary Study of OpenAI's Sora 2

Generative video models are increasingly capable of producing complex depictions of mental health experiences, yet little is known about how these systems represent conditions like depression. This study characterizes how OpenAI's Sora 2 generative video model depicts depression and examines whether depictions differ between the consumer App and developer API access points. We generated 100 videos using the single-word prompt "Depression" across two access points: the consumer App (n=50) and developer API (n=50). Two trained coders independently coded narrative structure, visual environments, objects, figure demographics, and figure states. Computational features across visual aesthetics, audio, semantic content, and temporal dynamics were extracted and compared between modalities. App-generated videos exhibited a pronounced recovery bias: 78% (39/50) featured narrative arcs progressing from depressive states toward resolution, compared with 14% (7/50) of API outputs. App videos brightened over time (slope = 2.90 brightness units/second vs. -0.18 for API; d = 1.59, q < .001) and contained three times more motion (d = 2.07, q < .001). Across both modalities, videos converged on a narrow visual vocabulary and featured recurring objects including hoodies (n=194), windows (n=148), and rain (n=83). Figures were predominantly young adults (88% aged 20-30) and nearly always alone (98%). Gender varied by access point: App outputs skewed male (68%), API outputs skewed female (59%). Sora 2 does not invent new visual grammars for depression but compresses and recombines cultural iconographies, while platform-level constraints substantially shape which narratives reach users. Clinicians should be aware that AI-generated mental health video content reflects training data and platform design rather than clinical knowledge, and that patients may encounter such content during vulnerable periods.

cs.CY↗

arXiv2025

Open-Source Full-Duplex Conversational Datasets for Natural and Interactive Speech Synthesis

Full-duplex, spontaneous conversational data are essential for enhancing the naturalness and interactivity of synthesized speech in conversational TTS systems. We present two open-source dual-track conversational speech datasets, one in Chinese and one in English, designed to enhance the naturalness of synthesized speech by providing more realistic conversational data. The two datasets contain a total of 15 hours of natural, spontaneous conversations recorded in isolated rooms, which produces separate high-quality audio tracks for each speaker. The conversations cover diverse daily topics and domains, capturing realistic interaction patterns including frequent overlaps, backchannel responses, laughter, and other non-verbal vocalizations. We introduce the data collection procedure, transcription and annotation methods. We demonstrate the utility of these corpora by fine-tuning a baseline TTS model with the proposed datasets. The fine-tuned TTS model achieves higher subjective and objective evaluation metrics compared to the baseline, indicating improved naturalness and conversational realism in synthetic speech. All data, annotations, and supporting code for fine-tuning and evaluation are made available to facilitate further research in conversational speech synthesis.

cs.SD↗

arXiv2025

Puzzle Game: Prediction and Classification of Wordle Solution Words

We study the prediction and classification of Wordle solution words. After cleaning the public results log, we fit an ARIMA model to forecast the daily volume of reported outcomes through March 1, 2023. For each solution word, we compute three interpretable attributes: usage frequency (FREQ), word information entropy (WIE), and the number of repeated letters (NRE), and analyze their correlations with the empirical attempt distribution (1-6 attempts plus failure, coded as 7). We then train an XGBoost regressor to predict the full 1-7 outcome distribution for unseen words; a case study of "EERIE" illustrates the model's behavior. To categorize difficulty, we cluster words into three tiers (simple, moderate, difficult) via K-means and train a decision-tree classifier that maps FREQ, WIE, and NRE to these tiers, yielding interpretable rules. For each word, we also report the share of players requiring three or more attempts. Sensitivity analyses and full modeling details are provided in the appendix.

math.NA↗