m-bain/whisperX

Convert arabic numerals and symbols to phonetic form

Open

#38 opened on Jan 25, 2023

View on GitHub
 (2 comments) (0 reactions) (0 assignees)Python (6,880 stars) (650 forks)batch import
help wanted

Description

Currently arabic numerals and symbols in whisper transcript cannot be aligned, needs to be phonetic alphabet.

Need to perform inverse of normalization in https://github.com/m-bain/whisperX/blob/main/whisperx/normalizers/english.py

Such that numbers and currencies are converted to their phonetic word form.

E.g. "$300" -> "three hundred dollars"

To perform wav2vec alignment.

Then convert back to symbol form, and assign timestamps.

Contributor guide

Tech stack
python
Domain
machine learningbackend
Issue type
feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
mostly clear
Prerequisites
familiarity with WhisperX codebaseunderstanding of text normalization
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
35
Research direction
The issue requires implementing the inverse of the normalization function in whisperx/normalizers/english.py to convert numbers and currencies into their phonetic word form (e.g., $300 > 'three hundred dollars') for wav2vec alignment. After alignment, the phonetic form must be converted back to symbolic form and timestamps assigned. The existing normalization function should be studied to understand the transformations. No linked PRs or maintainer comments provide further direction.