In drew-walkerr/musichistoRy: Functions To Compiling Last.fm User Music History With Spotify Audio Analysis

knitr::opts_chunk$set(echo = TRUE)
library(devtools)
#devtools::install_github('charlie86/spotifyr')
library(spotifyr)
library(tidyverse)
library(genius)
#devtools::install_github("drew-walkerr/musichistoRy")
library(musichistoRy)
library(tictoc)
library(syuzhet)
library(knitr)

tic()
myhistory <- get_history("thedrewwalker")
toc() 

tic()
myhistory_snip <- myhistory %>% 
  head(10) 
toc()

myaudiofeatures <- get_history_audio_features(myhistory_snip,"drew")

mysession_df <- get_session_df(myaudiofeatures)

#Should create group_by track dataset that summarizes playcount values and such 
unique_songs <- mysession_df %>% 
  distinct(id, .keep_all = TRUE)

#unique dataset for lyrics 
tic()
mylyrics <- unique_songs %>% 
  add_genius(artist, title, type = "track")
toc()
head(mylyrics, n=100)
kable(mylyrics)

Identifying Language and Translating Lyrics, building list of songs by language

• http://code.markedmondson.me/googleLanguageR/

install.packages("googleLanguageR")
library(googleLanguageR)
gl_auth("googletranslateR-717ce515dfc4.json")
tic()
mylyrics_detected <- mylyrics %>% 
  mutate(language = gl_translate_detect(lyric),
    en_translation = gl_translate(lyric))
toc()

mylyrics_detected2 <- flatten(mylyrics_detected)
save(mylyrics_detected, file = "mylyrics_detected.Rdata")

spanish_songs <- mylyrics_detected %>% 
  filter(language$language == "es")

spanish_songs_uniques <- spanish_songs %>% 
  distinct(artist,title)

Creating Anki decks

We'll transform the lyrics data to prepare for the deck import files, and utilize the reticulate package to interface with AnkiConnect to build Anki decks within R.

• AnkiConnect can build decks, and pull User Statistics, allowing for dashboard anki data as well

library(reticulate)

Using Python to generate decks

• https://github.com/kerrickstaley/genanki genanki may be better, looks like ankiconnect can only make blank decks

• https://medium.com/python-in-plain-english/make-flashier-flashcards-automating-anki-with-python-2744ed025366 looks like it could be helpful

```{python, get-anki-connect} import json import urllib.request

def request(action, **params): return {'action': action, 'params': params, 'version': 6}

def invoke(action, params): requestJson = json.dumps(request(action, params)).encode('utf-8') response = json.load(urllib.request.urlopen(urllib.request.Request('http://localhost:8765', requestJson))) if len(response) != 2: raise Exception('response has an unexpected number of fields') if 'error' not in response: raise Exception('response is missing required error field') if 'result' not in response: raise Exception('response is missing required result field') if response['error'] is not None: raise Exception(response['error']) return response['result']

invoke('createDeck', deck='Song') result = invoke('deckNames') print('got list of decks: {}'.format(result)) ```

Using syuzhet package for sentiment analysis

https://cran.r-project.org/web/packages/syuzhet/vignettes/syuzhet-vignette.html{.uri}

drew-walkerr/musichistoRy documentation built on May 29, 2021, 3:23 a.m.