Any specific reasoning to use the "lyric-api"? I do not directly know the Github project, nor have used it, however, from the documentation seems feasible to use. Database design also needs to be clarified.
Additionally, to note that there are legal limitations on crawling/scrapping websites, which need to be confirmed/checked in order to keep a fair use of the website to be crawled. This fact possibly requires adaptation of the crawling strategy as the data is collected, and thus of the deadline also to collect all data.
Regarding knowledge, I have previously crawled multiple websites using Node.js (e.g., phantomjs, puppeteer, apify).