crisishistory/HistoryAIToolkit

Create or find a smaller dataset of interview question and answer snippets

Open

#48 opened on Oct 4, 2023

View on GitHub
 (0 comments) (0 reactions) (0 assignees)Python (18 forks)github user discovery
hacktoberfest-acceptedhelp wantedhigh priority

Repository metrics

Stars
 (13 stars)
PR merge metrics
 (PR metrics pending)

Description

The current accompanying Kaggle dataset is a bit large for people on low bandwidth connections to download, and a bit large to annotate.

There are some existing speech recognition datasets, but I haven't seen one of questions and answers. Is there one? Or should we create our own?

  • Search online for a free speech recognition data set containing annotated, transcribed audio samples of questions and answers
  • If one exists, comment here with what you found
  • If one doesn't exist, create a tiny starter dataset consisting of 5 brief audio samples (1-3 minutes?) with accompanying transcription and timestamps. To help with this task, you can use slicer.py and transcript.py

Kaggle is one possible place to search for or host our own dataset. Open to other options too.

Contributor guide

Research direction
Look for existing speech recognition datasets containing Q&A snippets. If none exist, use slicer.py and transcript.py to create a small dataset of 5 audio samples with transcriptions and timestamps.
Tech stack
python
Domain
dataai
Issue type
Research
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-3 hours
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Fresh
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Mostly clear
Prerequisites
PythonGit
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
85