SpeechBubbleAI

Background

SpeechBubbleAI is the brainchild of a speech-language therapist who saw a major pain point in his profession: paperwork. Along with conducting therapy sessions, speech therapists have to spend a huge amount of time creating, filing, and searching through notes and progress reports. This admin work is not only tedious, but it also pulls focus away from what truly matters—their clients. The founder had a vision to change that by harnessing AI to help therapists spend more time with patients and less time on paperwork.

Solution

We partnered with SpeechBubbleAI to build a proof-of-concept tool that would lighten the load for therapists. Using Astro for the front end and PostgresDB for the back end, we created a simple web app. Here’s how it works: therapists can upload their session transcripts (previously imported from Zoom) into the app. Once uploaded, the transcripts are processed by Claude, a large language model (LLM), which was prompted to analyze the conversation and generate a SOAP (Subjective, Objective, Assessment, Plan) note. The generated note serves as a draft that the therapist can quickly review, edit, and store for future reference.

Results

The project is still in its early stages, and the team is actively collecting feedback and testing the market to determine if there’s a strong demand for this tool. While initial reactions are promising, the real impact on therapists’ workflows is still being evaluated. As testing continues, we’ll gather data and analyze whether the tool truly helps streamline the note-taking process and improves overall efficiency. Exact numbers and measurements will be available once the testing phase is complete.