Seamlessly shorten songs. No cut ends! Live at https://projects.loud.red/snipsnip/
This is a web service that allows users to shorten a song without cutting off the end. Instead, similar portions of audio are identified and cross-faded to produce a new song that retains the start and end of the original song with a middle portion removed.
The front-end is a static website hosted on S3 and communicates with endpoints on AWS that call Python functions run on Lambda to process files. The serverless framework is used to make deployment easy. gulp is used to build the front-end (compiling the ES6 code and Sass files and minifying them for production).
Make any changes you need to, make sure serverless is set up and run npm run deploy. You may also find the following commands helpful:
npm run serveto run an HTTP server (for front-end work)npm run devto recompile Sass files when changednpm run buildto build the front-endnpm run deployto build and upload the entire project to AWSnpm run deploy-webto build and update just the static website
- slimmed-down copy of pyAudioAnalysis
- custom Zepto build (using
MODULES="zepto event ie" ./make dist) - Elusive Icons built using Fontello
- ffmpeg builds for Lambda