Author: Jonathan Lim Wei Siang Email: Jonathanlimws@gmail.com
# Main Function This Python package allows you to assess the phonetic error rate and visualise them.
# Workflow of package 1. TIMIT file is loaded 2. ASR Model converts audio.wav file into phoneme string 3. ASR Phoneme string is standardized to TIMIT standard to allow for comparison 4. Phonetic Error Rate is generate
## Additional Noise can be added to audio.wav fiel to test the ASR model and step 2-4 is repeated
# Types of Plots 1. Boxplot of accuracy rate for each phoneme across selected TIMIT files 2. Stacked boxplot of accuracy rate across varying added noise 3. Time/frequency Plot any given TIMIT audio showing the timing/phoneme which was incorrected predicted (substitution and deletion only) 
# Requirements TIMIT file: TIMIT is a corpus of phonemically and lexically transcribed speech of American English speakers of different sexes and dialects link: https://en.wikipedia.org/wiki/TIMIT
Change Log
0.0.1 (12/08/2022)
First Release
1 maintainer
Metadata
1 maintainer
Release files for asr-phoneme-assessment 0.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| asr_phoneme_assessment-0.0.1.tar.gz | 13.7 kB | Details |
Release files / asr_phoneme_assessment-0.0.1.tar.gz
| Download URL | asr_phoneme_assessment-0.0.1.tar.gz |
|---|---|
| Size | 13.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
2ab7804b441d3fd8929170cd3ace1fd5e2b0a577e31be2c1f82b2c140851187f
|
|
BLAKE2b-256 checksum How to use checksums |
4fe6bc31dbb2454169945f03dac53ebf34c370eed1da27856e7ee43bfe319ff0
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.1 CPython/3.9.6
|
