Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Emotion Recognition among Couples
At the Center for Digital Health Interventions at ETH Zurich, we seek to develop a smartwatch-based app to recognize the emotions of couples in daily life to be used for research on couples’ emotions in everyday life and enable the development of interventions to improve their emotional well-being.
Keywords: Affective Computing, Wearable Computing, Mobile Health, Machine Learning, Deep Learning
We are offering bachelor’s and master’s theses opportunities related to this project. We have two datasets for analysis:
1) Speech data collected from Dutch-speaking couples in Belgium (Lab data)
2) Smartwatch data (audio, heart rate, accelerometer, and gyroscope) collected from German-speaking couples in Switzerland (Field data)
We are offering bachelor’s and master’s theses opportunities related to this project. We have two datasets for analysis:
1) Speech data collected from Dutch-speaking couples in Belgium (Lab data)
2) Smartwatch data (audio, heart rate, accelerometer, and gyroscope) collected from German-speaking couples in Switzerland (Field data)
You will build data analytics pipelines entailing data preprocessing, feature extraction, developing, and training machine models for emotion recognition which will result in papers to be published in venues such as UbiComp, ICMI, Interspeech, ICASSP, etc.
**Potential Bachelor’s/Master’s Thesis Topics**
- Analysis of couples’ interactions in everyday life using sensor data
- Multimodal emotion recognition among couples using real-world data from couples
- Emotion Recognition among Couples using the Peak, End and Whole Audio
- Speaker Diarization among Couples using Gestures from Smartwatches
- Continuous emotion recognition using Couples' Speech data
- Predicting Couples’ End-of-Conversation Emotion using Interaction Dynamics
- Emotion recognition among Older Adults using Real-World Speech Data
**Skills Requirement**
Programming (Python) and Applied Machine Learning and or Deep Learning experience
You will build data analytics pipelines entailing data preprocessing, feature extraction, developing, and training machine models for emotion recognition which will result in papers to be published in venues such as UbiComp, ICMI, Interspeech, ICASSP, etc.
**Potential Bachelor’s/Master’s Thesis Topics**
- Analysis of couples’ interactions in everyday life using sensor data - Multimodal emotion recognition among couples using real-world data from couples - Emotion Recognition among Couples using the Peak, End and Whole Audio - Speaker Diarization among Couples using Gestures from Smartwatches - Continuous emotion recognition using Couples' Speech data - Predicting Couples’ End-of-Conversation Emotion using Interaction Dynamics - Emotion recognition among Older Adults using Real-World Speech Data
**Skills Requirement** Programming (Python) and Applied Machine Learning and or Deep Learning experience
George Jojo Boateng
If you are interested, kindly send an email with your **CV, transcript, and a motivation statement** in which you mention **which thesis topic(s) interest you**.
Learn more about the project here: https://www.c4dhi.org/projects/dyadic-management-diabetes/
George Jojo Boateng
If you are interested, kindly send an email with your **CV, transcript, and a motivation statement** in which you mention **which thesis topic(s) interest you**.
Learn more about the project here: https://www.c4dhi.org/projects/dyadic-management-diabetes/