Finding therapists who accept insurance can be a daunting task. Psychologists, social workers, and psychiatrists who don't accept insurance say that insurers' reimbursement rates are too low. [1] Ethnic minorities struggle further to find a culturally competent therapist.

What it does

Therapeer matches patients with therapists using sentiment analysis, along with their demographic information. First, the patient records their feelings and thoughts. This recording is analyzed and sent back to the patient with results. which display the individual's emotion and, based on the sentiment analysis, receive a match with a therapist who could help. The patient has the option to call or chat with the therapist to provide both, synchronous and asynchronous modes of treatment. Additionally, Therapeer uses micro-transactions to let the patient pay for their therapy session dependent on the session's duration, rather than a fixed rate. This can significantly reduce the expenses incurred when paying out of pocket.

Key Features

  • Sentiment analysis of both text and audio from a call/recording
  • Uses A Deep learning model to classify emotions/sentiment in recorded human speech audio
  • Uses assembly AI to transcribe the recordings with speaker diarization
  • Uses Google Cloud Natural Language Processing to determine textual sentiment
  • Uses micropayments to pay therapists
  • Micropayments are implemented as part of an ERC 20 custom token called TherapeerCoin (TPC) Contract address is at 0x74E1a11806CCFd8b768dDDeB2615668d903C79d7

How we built it

  • React Native
  • Assembly AI
  • Google Cloud Platform
  • Python and Flask Framework

Challenges we ran into

Sentiment analysis was quite a challenge

Accomplishments that we're proud of

It works!

What we learned

Therapy can be crazy expensive!

What's next for Therapeer

Onboard therapists for pro-bono sessions


Built With

Share this project: