We're researching things at the edge of of voice science - from diagnosing diseases with new machine learning models to building open-source voice assistants.
We're creating new protocols for how to collect and clean voice data.
We're making new feature embeddings for voice, text, image, and video data.
We're curating new speech datasets to classify emotions, accents, gender, and age.
We're looking at new ways we can automatically generate text and voice data.
We're innovating the process of visualizing audio, text, and image data with Elastic Search and Kibana.
We've open sourced our own voice assistant, Nala, written completely in Python.
We're researching new distributed computing server paradigms for voice computing.
We're researching new security, legal, and ethics frameworks to maximize consumer protections.
We're an interdisciplinary group of engineers, computer scientists, physicians, lawyers, designers, and researchers working to translate work in voice computing into products.
1. Parkinson's detection using raw data from the iphone accelerometer.
2. Validating methods for early detection of PD symptoms with voice.
3. Closed-loop feedback deep brain stimulation project for people with PD/tremor.
4. Depression detection among those with PD.
5. Depression detection from the Audio-Visual Emotion Challenge.
6. Automated Voice Biomarkers for Depression Symptoms using an Online Cross-Sectional Data Collection Initiative.
7. Voice Biomarkers for Alzheimer’s Disease: Predicting Neuropsychological Measures and Dementia Status in the Framingham Heart Study Cognitive Aging Cohort. [coming soon].
8. Schwoebel, J. (2018). An Introduction to Voice Computing in Python. Boston; Seattle; Atlanta: NeuroLex Laboratories. (150+ GitHub stars).
9. A voice analysis pipeline to help speed up research in vocal biomarkers.
10. Villongco, Christopher, and Fazal Khan. "Sorry I Didn’t Hear You.” The Ethics of Voice Computing and AI in High Risk Mental Health Populations." AJOB neuroscience 11.2 (2020): 105-112.
Group 1 (1:50-13:44) - SES Biomarker
Yosef Amrami (14:45-23:50) - Appraisals
Jordan Lander (27:00-43:30) -Alcohol
Prateek Gupta (44:22-1:01:47) - Scarce data
Nikhil Raghuraman (1:03:17-1:13:00) - Twilio
Jerry Shan - (1:14:00-1:23:00) - iOS framework
Agnes Zhao - (1:24:00-1:31:00) - YouTube
Katy Felkner - (1:31:20-1:39:32) - Dementia
Maria Daigle (0:00-0:10) - Stress
Haoxin Li (10:00-20:00)- Accents
Raj Singh (20:00-30:00) - Ages
Tarun Maddali (30:00-40:00) - Stress
Dylan Martin (50:00-1:00:00) - Laserbeak
Kathleen Williams (1:00-1:10) - Depression
Justin John / Annie Vu (1:10:00) - Autism
Kamilah Mitchell (1:20:00) - Formant extraction
Luke Lyon (video link) - Emotions
Hilary Lynch (video link)- Alzheimer's
Tim Wroge (0:00-13:20) - Parkinson's disease
Larry Zhang (13:20-20:25) - Data pipelines
Audrey Wagner (20:25-33:29) - Glioblastoma
Dylan Pitulski (33:29-51:18) - Schizophrenia
Wendy Nguyen (53:00-1:03:36) - Depression
Allison Pei (1:03:00-1:13:00) - Common colds
Alyssa Naritoku (1:16:00-1:21:00) - Graves' disease
Mugdha Apte (1:22:00-1:30:00) - Anxiety disorders
Peter Tang (0:00-11:40) - Caffeine Detection
Shadab Hassan (11:40-23:00) - Anxiety
Alice Romanav (23:40-35:30) - Parkinson's
Radhika Duvvuri (39:29-48:44) - Depression
Sumaiya Sayeed (53:17-1:02:50) - Addiction
Yahia Ali (1:36:00-1:49:00) - Hardware demo
Jake Peacock (1:55:00-2:08:00) - Elastic Search
Hack Anxiety is the first Hack-X format that we hosted at Georgia Tech. The idea is to have a 3 hour Hack-a-thon format for innovators to come up with solutions in voice computing related around real problems - in this case, diagnosing and managing anxiety disorders.