https://github.com/capjamesg/taylor-swift
Find how similar your voice is to Taylor Swift (WIP) ✨
Science Score: 13.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
✓codemeta.json file
Found codemeta.json file -
○.zenodo.json file
-
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (12.8%) to scientific vocabulary
Keywords
Repository
Find how similar your voice is to Taylor Swift (WIP) ✨
Basic Info
- Host: GitHub
- Owner: capjamesg
- Language: Python
- Default Branch: main
- Homepage: https://swifties.me
- Size: 588 KB
Statistics
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
- Releases: 0
Topics
Metadata Files
README.md
swifties.me
swifties.me is a web application that lets you find out how similar your voice is to Taylor Swift. Swifties offers two voice comparison methods: (i) using ImageBind audio embeddings, and; (ii) using the SpeechBrain speaker verification model.
More investigation is needed to determine which method is more accurate. It is not clear whether ImageBind is good for speaker identification: in testing, qualitatively, ImageBind ranked Taylor Swift speaking higher than non-Taylor Swift samples. But, there were false positives. This was the case for SpeechBrain. It may be the case that another approach would achieve better results.
How it Works
Both the ImageBind and SpeechBrain approach accept an arbitrary number of clips of Taylor Swift singing. These clips need to have isolated vocals so that they represent Swift singing rather than the audio in the background. [Demucs] by Meta Research proved effective in this task.
When a user records an audio clip, it is compared against either:
- Pre-computed ImageBind embeddings for the isolated vocal clips, or;
- The raw vocal clips using the SpeechBrain speaker verification model.
A similarity score is then returned to the user.
Getting Started
ImageBind
First, you will need to install ImageBind. You can do this by following the official ImageBind installation instructions.
In the ImageBind root folder, copy the web app in this repository (image_bind.py). Then, install all the requirements for this project:
bash
pip install -r ts-requirements.txt
Then, run the web app in debugging mode:
bash
python3 web.py
SpeechBrain
First, install the dependencies for the project:
bash
pip install -r ts-requirements.txt
Then, run the web app in debugging mode:
bash
python3 web.py
Using the App
After running the web.py script, the web application will be available at localhost:8084. You will need to run the app on a HTTPS server as browser audio recording requires a secure context.
Experiments
The experiments folder contains miscellaneous code not used in the application.
License
This project is licensed under an MIT license.
Acknowledgements
- The HTML and CSS for this app were started in Glitch. Thank you Glitch for making an intuitive web interface for HTML editing.
- ImageBind and SpeechBrain for their open source projects.
Owner
- Name: James
- Login: capjamesg
- Kind: user
- Location: Scotland
- Company: @Roboflow
- Website: jamesg.blog
- Repositories: 320
- Profile: https://github.com/capjamesg
from words, wonder.
GitHub Events
Total
Last Year
Issues and Pull Requests
Last synced: 11 months ago
All Time
- Total issues: 0
- Total pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Total issue authors: 0
- Total pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Past Year
- Issues: 0
- Pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Issue authors: 0
- Pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0