https://github.com/web-dev-codi/transcript-productivity-parser
A Python application that automatically extracts actionable tasks from transcript files and converts them into organized daily todo lists in CSV format. Perfect for developers who record meetings, brainstorming sessions, or voice notes and want to quickly identify and track follow-up tasks.
https://github.com/web-dev-codi/transcript-productivity-parser
csv microsoft parser python teams todo todolist transcript
Last synced: 4 months ago
JSON representation
A Python application that automatically extracts actionable tasks from transcript files and converts them into organized daily todo lists in CSV format. Perfect for developers who record meetings, brainstorming sessions, or voice notes and want to quickly identify and track follow-up tasks.
- Host: GitHub
- URL: https://github.com/web-dev-codi/transcript-productivity-parser
- Owner: Web-Dev-Codi
- Created: 2025-07-23T12:51:52.000Z (6 months ago)
- Default Branch: main
- Last Pushed: 2025-07-23T14:05:49.000Z (6 months ago)
- Last Synced: 2025-07-23T15:15:20.095Z (6 months ago)
- Topics: csv, microsoft, parser, python, teams, todo, todolist, transcript
- Language: Python
- Homepage:
- Size: 13.7 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Transcript Todo Parser
A format-agnostic Python tool that extracts todo items and action items from Microsoft Teams meeting transcripts and exports them to CSV format.
## Features
- **Format Agnostic**: Supports VTT, SRT, TXT, and JSON transcript formats
- **Intelligent Extraction**: Identifies todos using multiple keyword patterns and context analysis
- **Speaker Detection**: Automatically extracts speaker names from transcripts
- **Assignment Recognition**: Identifies who is assigned to each todo item
- **CSV Export**: Outputs structured data with timestamp, speaker, todo text, and assignee
- **CLI Interface**: Easy-to-use command line interface with preview mode
## Installation
1. Create and activate a virtual environment:
```bash
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
```
2. Install required dependencies:
```bash
pip install -r requirements.txt
```
## Usage
### Basic Usage
```bash
python main.py transcript.vtt
python main.py meeting_notes.txt -o todos.csv
```
### Preview Mode (without saving)
```bash
python main.py transcript.txt --preview
```
### Verbose Output
```bash
python main.py transcript.vtt --verbose -o detailed_todos.csv
```
## Supported Keywords
The parser automatically detects todos when speakers use these keywords:
- "todo", "action item"
- "need to", "have to", "must do"
- "will do", "should do", "going to"
- "will handle", "will work on", "will take care of"
- "responsible for", "assigned to"
## Output Format
The CSV output includes these columns:
- **timestamp**: When the todo was mentioned
- **speaker**: Who mentioned the todo
- **keyword**: The trigger keyword found
- **todo**: The extracted todo text
- **assignee**: Who is assigned (if identified)
- **context**: Full context of the statement
## Examples
Run with the provided sample files:
```bash
python main.py example_transcript.txt --preview
python main.py example_teams.vtt -o sample_output.csv
```
## Supported Formats
- **VTT/WebVTT**: Microsoft Teams and other video platform exports
- **SRT**: Subtitle format files
- **TXT**: Plain text transcripts with speaker identification
- **JSON**: Structured transcript data