pilot7747/VoxDIY

This repository provides data and code for "Vox Populi, Vox DIY: Benchmark Dataset for Crowdsourced Audio Transcription" paper.

27
/ 100
Experimental

This project provides datasets and tools for evaluating and improving crowdsourced audio transcriptions. It helps you assess the quality of transcripts submitted by multiple workers for the same audio files, and can help you create robust new audio datasets with validated human input. Data consists of audio files, individual crowd worker transcriptions, and their corresponding ground-truth texts. It's intended for researchers and practitioners working on speech-to-text technologies or managing crowdsourcing annotation projects.

No commits in the last 6 months.

Use this if you need to evaluate the accuracy of transcriptions from a crowd of workers against a known 'ground truth' or synthesize speech for creating new transcription tasks.

Not ideal if you're looking for a simple, out-of-the-box solution to transcribe audio without involving a crowdsourcing workflow or human quality control.

audio-transcription crowdsourcing speech-to-text data-annotation quality-control
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 5 / 25

How are scores calculated?

Stars

16

Forks

1

Language

Python

License

Last pushed

Jul 22, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/pilot7747/VoxDIY"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.