Detecting Depression with Audio/Text Sequence Modeling of Interviews

@inproceedings{Hanai2018DetectingDW,
  title={Detecting Depression with Audio/Text Sequence Modeling of Interviews},
  author={Tuka Al Hanai and Mohammad Mahdi Ghassemi and James R. Glass},
  booktitle={INTERSPEECH},
  year={2018}
}
Medical professionals diagnose depression by interpreting the responses of individuals to a variety of questions, probing lifestyle changes and ongoing thoughts. [...] Key Method We utilized data of 142 individuals undergoing depression screening, and modeled the interactions with audio and text features in a Long-Short Term Memory (LSTM) neural network model to detect depression. Our results were comparable to methods that explicitly modeled the topics of the questions and answers which suggests that…Expand
Predicting Depression in Screening Interviews from Latent Categorization of Interview Prompts
TLDR
JLPC, a model that analyzes interview transcripts to identify depression while jointly categorizing interview prompts into latent categories allows the model to define high-level conversational contexts that influence patterns of language in depressed individuals. Expand
Detecting Depression with Word-Level Multimodal Fusion
TLDR
This paper proposes a model that is able to perform modality fusion incrementally after each word in an utterance using a time-dependent recurrent approach in a deep learning set-up, and uses fusion gates that control the degree to which the audio or visual modality contributes to the final prediction. Expand
Identifying Mood Episodes Using Dialogue Features from Clinical Interviews
TLDR
This work moves beyond acoustic and lexical information, investigating how higher-level interactive patterns also change during mood episodes and shows that it is beneficial to consider dialogue features when analyzing and building automated systems for predicting and monitoring mood. Expand
Evaluating Acoustic and Linguistic Features of Detecting Depression Sub-Challenge Dataset
TLDR
In this analysis, key factors to consider during pre-processing and modelling to effectively build voice biomarkers for depression are pointed out and the dataset is verified for balance in demographic and severity score distribution. Expand
Text-based Decision Fusion Model for Detecting Depression
TLDR
A decision fusion model is proposed which combines Bert text embedding of interview transcript and key phrases recognition which outperforms previous single-modality approaches in terms of classification accuracy. Expand
Detecting Depression in Less Than 10 Seconds: Impact of Speaking Time on Depression Detection Sensitivity
TLDR
Results show that 10 seconds or less are sufficient to achieve the same level of recall (roughly 70%) observed after using the entire inteview of every participant, while reducing by 95%, on average, the amount of time required to collect the necessary data. Expand
Hybrid approach to detecting symptoms of depression in social media entries
TLDR
This study presents an innovative approach to the depression screening problem by applying Collgram analysis, which is a known effective method of obtaining linguistic information from texts, and creates a hybrid model achieving a diagnostic accuracy of 71%. Expand
The Detection of Depression Using Multimodal Models Based on Text and Voice Quality Features
The article proves the concept that an automatic diagnosis of depression can be achieved using audio recordings of the individuals’ voices. DAIC-WOZ database was used as a data source. Audio andExpand
A Neural Network Based Hybrid Model for Depression Detection in Twitter
Depression is a serious mental illness that leads to social disengagement, affects an individual’s professional and personal life. Several studies and research programs are conducted forExpand
Screening For Depression With Retrospectively Harvested Private Versus Public Text
TLDR
This is the first study to leverage the retrospectively-harvested crowd-sourced texts and tweets within the combined Moodable and EMU datasets to predict scores from the Patient Health Questionnaire-9, a prevalent depression screening instrument. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 25 REFERENCES
Topic Modeling Based Multi-modal Depression Detection
TLDR
This work proposes a novel topic modeling based approach to perform context-aware analysis of the recording of an interview and shows that the proposed approach outperforms context-unaware methods and the challenge baselines for all metrics. Expand
Detecting Depression using Vocal, Facial and Semantic Communication Cues
Major depressive disorder (MDD) is known to result in neurophysiological and neurocognitive changes that affect control of motor, linguistic, and cognitive functions. MDD's impact on these processesExpand
Decision Tree Based Depression Classification from Audio Video and Language Information
TLDR
A decision tree is constructed according to the distribution of the multimodal prediction of PHQ-8 scores and participants' characteristics obtained via the analysis of the transcript files of the participants to improve the recognition accuracy of the Depression Classification Sub-Challenge (DCC) of the AVEC 2016. Expand
A Random Forest Regression Method With Selected-Text Feature For Depression Assessment
TLDR
A random forest method with a Selected-Text feature based method which is according to the analysis on the transcript in different depressive levels has obtained very promising results on the development and test sets. Expand
The Distress Analysis Interview Corpus of human and computer interviews
The Distress Analysis Interview Corpus (DAIC) contains clinical interviews designed to support the diagnosis of psychological distress conditions such as anxiety, depression, and post traumaticExpand
Hybrid Depression Classification and Estimation from Audio Video and Text Information
TLDR
Results show that the proposed depression recognition framework obtains very promising results, with root mean square error (RMSE) as 3.088, mean absolute error (MAE) as 2.477 on the development set, and RMSE as 5.359 on the test set, which are all lower than the baseline results. Expand
Vocal biomarkers of depression based on motor incoordination
TLDR
This paper designs a novel Gaussian mixture model (GMM)-based multivariate regression scheme, referred to as Gaussian Staircase Regression, that provides measures of coordination in vocal tract articulation while reducing effects of a slowly-Varying linear channel, which can be introduced by time-varying microphone placements. Expand
Depression Severity Prediction Based on Biomarkers of Psychomotor Retardation
TLDR
This paper offers a novel methodology to target specific craniofacial movements indicative of psychomotor retardation and hence of depression and presents a novel method for quantifying abnormalities of speech spectra of individuals with depression using Fisher vector encoding of spectral low level descriptors. Expand
Depression Assessment by Fusing High and Low Level Features from Audio, Video, and Text
TLDR
The present paper reports on the results of the participants' participation to the depression sub-challenge of the sixth Audio/Visual Emotion Challenge (AVEC 2016), which was designed to compare feature modalities in gender-based and gender-independent modes using a variety of classification algorithms. Expand
DepAudioNet: An Efficient Deep Model for Audio based Depression Classification
TLDR
A deep model is proposed, namely DepAudioNet, to encode the depression related characteristics in the vocal channel, combining Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) to deliver a more comprehensive audio representation. Expand
...
1
2
3
...