Copyright © 2025 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{156039,
        author = {Prajwal B and Ghanavi Yadav A and Divyashree P and Namratha Jayadev and Sharmila Chidaravalli},
        title = {Stress and Anxiety Detection through Speech Recognition and Facial Cues using Deep Neural Network},
        journal = {International Journal of Innovative Research in Technology},
        year = {},
        volume = {9},
        number = {2},
        pages = {578-582},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=156039},
        abstract = {Stress is a feeling of emotional tension and anxiousness. It can have an influence on each individual’s mental health. On the other hand, anxiety is a common reaction to stress which makes one fearful thus leads to panic attacks. Stress and anxiety can lead to unreasonable complications with an individual’s personal life.  These mental issues can cause mental instability which has to be treated in a right manner. This paper analyses how we use vocal/audio dataset and video visuals i.e. facial expressions to detect stress and anxiety in an individual. Here we have developed a model where stress and anxiety is detected using deep neural network. We use these actors audio/vocal datasets from Kaggle where the audio consists of 7 emotions i.e., anger, surprised, sadness, neutral, disgust, fear and joy. Later the audio datasets are used to train and test few of the classification models like Convolution Neural Network (CNN). Then the audio which is collected will be pre-processed through acoustic feature extraction, accordingly the audio is classified through CNN which provides the accuracy based on those 7 emotions. CNN is also applied to analyze visuals where it derives a relationship between pixels by determining features of an image using input data. Input image is passed through convolution layers with filters like kernel to produce the outcome of facial expressions. By this we can predict if the person has stress or anxiety.},
        keywords = {Convolutional Neural Network, Emotion Classification, Stress Detection, MFCC (Mel frequency cepstral coefficients), Chroma.},
        month = {},
        }