Simple audio file labeling tool for speech diarization/recognition.
Allows users to analyze audio files and manually label speakers, which is parsed into a text file. Developed inorder to train our "Speech Diarization" machine learning model.
Features:
- Label up to 3 speakers
- Fastforward/Rewind effiently
- Quickly access to all audio files
-
Drop all .wav/.mp3 files into 'Audio' File
-
Download Offline web server below and set directory to 'Audio' File (https://chrome.google.com/webstore/detail/web-server-for-chrome/ofhbbkphhbklhfoeikjpcbhemlocgigb?hl=en)
HINT: Local URL should be http://127.0.0.1:8887/. If not, make approriate changes in index.js */
-
Download 'Allow-Control-Allow-Origin' extention to allow local file access (https://chrome.google.com/webstore/detail/allow-control-allow-origi/nlfbmbojpeacfghkpbjhddihlkkiljbi)
-
Open index.html
Created by: Krishna Karunaharan