Skip to main content

Analyzing Audio Files to Determine Therapy Efficacy

Posted by on Wednesday, January 19, 2022 in Uncategorized.

Current Project: Using automatic speech recognition to transcribe audio files from patients with depression to determine efficacy of therapy. Before and after outcomes will be assessed using telephone audio files where participants were called and asked what they were “thinking” at the moments just before the call. These audio files can be coded using audio and text sentiment analysis.

Future project 1: Using automatic speech recognition to transcribe audio files that are sampled from 12-30 second audio clips that are sampled from smart watches over the course of one-to-two week 8 hour days. These files will be associated with experience sampling data (over the same time periods) that correspond to questions about people’s personality. I’m hoping to use these audio sound bites to contextualize personality.

Future project 2 (more ambitious in scope): Using automatic speech recognition (to transcribe to text) or text (from participant’s own typing) to answers to the question of “Who are you?” and “How would you describe your personality?” These descriptions can then be used to generate personalized personality items (e.g., Hommel et al., 2021). The purpose would be to embedded these questions on an open-source website (e.g., where people can come and fill out these questions as they please. Using these transformers, we can create customized personality questions that can be provided to participants in daily surveys that are delivered to their phone via a web link (e.g., using formr or Vanderbilt’s RedCap). The overarching goal would be to have an automated pipeline for data collection of personality that is personalized to the person that does not need researcher direction (i.e., the data collection is automated, creating an open-source database of personalized personality). People that participate would be provided with feedback at the end of their participation.

If you have any questions about this project and/or are interested in joining the team, please reach out to