Skip to content
Samprat Trivedi

Work

How they connect

These three projects weren't planned as a set. Each one started with a question the previous one left me with.

It began with Wispr Flow, an app I use to dictate instead of type. I liked it, and wondered why more people don't use ChatGPT's voice mode. I assumed they couldn't find it. My research showed something else: many people felt awkward talking to their phone around others.

Interviewing people for that research, I got more interested in their frustrations than in voice itself. Product teams get thousands of reviews and complaints, so how do they make sense of them? That became VoC Copilot.

VoC Copilot used AI to sort the reviews, which raised the next question: how much should a team trust what AI tells it? GrowPulse, a project for my NextLeap cohort, was my answer. The model only sorts; code owns every number, quote and link.

  1. 1Wispr FlowDictating instead of typing
  2. 2Voice mode researchWhy don’t more people use it?
  3. 3VoC CopilotHow do teams make sense of feedback?
  4. 4GrowPulseHow far should we trust AI insights?
How each project led to the next.
ChatGPT voice mode adoption in IndiaFellowship research

Most people who skip ChatGPT’s voice mode aren’t blocked by anything. The ones who use it go quiet in public, so I prioritised a whisper mode.

36 survey responses. 5 one-to-one interviews. 3 hypotheses tested. 4 solutions ranked by effort and impact.

Starting thesis
  • Low usage is mainly a discoverability problem
What the research showed
  • Non-users can find it, they prefer typing
  • Users go quiet when others are around
  • Mixed-language speech is misheard
The thesis I started with, and what 36 survey responses and five interviews pointed to instead.

Read the case study