TranscriptX

TranscriptX

Think with your transcripts — on your machine.

Import conversations you already have. See themes, speakers, and evidence in one place. Optional local AI stays on your computer.

See how it works Install

The application

Home and recent runs, name speakers, fix the transcript, run analysis, then read Overview, the transcript, Insights, and Charts.

Home: corpus totals and recent runs
Home — corpus totals and recent runs
Speaker Identification: diarized speakers named from their lines
Speaker Identification — name diarized speakers
Corrections Studio: review candidates and generate from the transcript
Corrections Studio — review and fix transcript text
Run Analysis: Balanced preset and custom questions from the library
Run Analysis — Balanced preset and custom questions
Overview after a Balanced run: summary and at-a-glance metrics
Overview — summary and at-a-glance metrics
Transcript view with named speakers, timestamps, and search
Transcript — named speakers, timestamps, and search
Insights Highlights: notable moments with quoted lines
Insights Highlights — notable moments with quotes
Charts gallery for the finished run
Charts — gallery for the finished run

What can I do with it?

  • Understand themes across a conversation
  • Compare speakers — who said what, and how they interact
  • Investigate a question and jump back to the original lines
  • Analyse several conversations together over time
  • Correct the transcript while you read
  • Export findings as HTML or a ZIP you keep

Explore the workflows for step-by-step jobs, or see how it compares to transcription tools, meeting assistants, and research software.

On your machine

Source files and analysis results stay on your computer. Optional AI uses Ollama locally and stays off until you turn it on. Nothing here is a hosted analysis service.

From a file to a useful Overview

  1. Open Import Transcript and upload a JSON file.
  2. Open Speaker Identification and give each diarized label a display name. Most Balanced modules skip until speakers are named.
  3. Open Run Analysis, keep Balanced, and run it.
  4. Open Overview and read the summary and the at-a-glance metrics.

Full walkthrough with screenshots · All workflows

Install

Docker is the simplest path. Copy .env.example to .env and set HOST_RECORDINGS_DIR to a folder outside the repository. 

git clone https://github.com/glen-w/TranscriptX.git
cd TranscriptX
cp .env.example .env   # set HOST_RECORDINGS_DIR
docker compose up transcriptx-web

Open http://localhost:8501. The first run builds the image.

Native install and troubleshooting: Installation · Bring transcript files · README