Aug 26, 2020

Dr. Jessica Robin is a scientist at Winterlight Labs. Winterlight Labs is a speech analytics company based in Ontario, Canada. They use applied healthcare technology to assess Alzheimer’s and dementia progression. In this episode, Dr. Jessica Robin discusses how voice samplings can be used to make predictions about diseases. 

Top three takeaways: 

1. Dr. Jessica Robin and her team give patients a picture to look at, and the patient describes what they see, giving them speech data to analyze.
2. Current projects include mining interviews of celebrities for data showing alterations in speech as a prediction to Alzheimer’s. 
3. The data collected ranges from unstructured and structured speech to assess the acoustics, pacing, and syntax. 

[0:00] Ladan introduces the episode and the guest, Dr. Jessica Robin.  

[2:50] Just one minute of speech can give over 500 variables of speech data to analyze. 

[3:40] This type of technology can be applied to presidential speeches, and studies show that it is useful in observing healthy versus Alzheimer’s patients’ voices. 

[4:50] While certain conditions of voice samples such as previously prepared speeches do not provide the choice of syntax, acoustic and cadence data can be observed instead. 

[7:00] Dr. Jessica Robin wishes to expand into psychiatric disorders since aspects of a person’s mental state are reflected in how they speak.