Last active
December 19, 2015 15:28
-
-
Save kl/5976027 to your computer and use it in GitHub Desktop.
Yotsubato TTS downloader
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#encoding: utf-8 | |
require 'fileutils' | |
require 'open-uri' | |
require 'sanitize' | |
#require 'pry' | |
class YotsubatoDownloader | |
API = ARGV[1] || "http://208.109.168.116/GetAudio1.ashx?speaker=302&content=" | |
SAVE_FOLDER = "./tts/" | |
OUT_NAME = "import_me.txt" | |
SEPARATOR = "\t" | |
def initialize | |
@hash = Time.now.hash.abs.to_s | |
end | |
def download(file_path) | |
reader = NotesReader.new | |
items = reader.read_items(file_path) | |
make_out_file(items) | |
download_wavs(items) | |
end | |
private | |
def make_out_file(items) | |
File.open(OUT_NAME, "w") do |out_file| | |
items.each_with_index do |item, index| | |
trans = item[:english] + " [sound:#{@hash}_#{index}.wav]" | |
if item[:kanji] | |
line = item[:kanji] + SEPARATOR + item[:kana] + SEPARATOR + trans | |
else | |
line = item[:kana] + SEPARATOR + " " + SEPARATOR + trans | |
end | |
out_file.puts(line) | |
end | |
end | |
end | |
def download_wavs(items) | |
make_save_dir | |
items.each_with_index do |item, index| | |
wav = get_wav_data(item[:kana]) | |
write_wav(wav, "#{@hash}_#{index}.wav") | |
puts "Downloading .wav #{index}" | |
end | |
end | |
def make_save_dir | |
FileUtils.mkdir(SAVE_FOLDER) unless File.directory?(SAVE_FOLDER) | |
end | |
def get_wav_data(japanese) | |
request = API + URI::encode(japanese + "。") | |
open(request).read | |
end | |
def write_wav(wav_data, filename) | |
File.open(SAVE_FOLDER + filename, "wb") { |f| f.write(wav_data) } | |
end | |
end | |
class NotesReader | |
def read_items(file_path) | |
IO.foreach(file_path).map do |line| | |
cleaned = Sanitize.clean(line) | |
fields = cleaned.split("\t") | |
reading = fields[0].lstrip.rstrip | |
rest = fields[1].lstrip.rstrip | |
if rest.include?(" ") | |
kana, english = rest.split(" ") | |
{kanji: reading, kana: kana, english: english} | |
else | |
{kanji: nil, kana: reading, english: rest} | |
end | |
end | |
end | |
end | |
##### SCRIPT START ##### | |
if ARGV[0].nil? | |
puts "Usage: yotsubato.exe [path to exported file]" | |
exit | |
end | |
html_file = File.expand_path(ARGV[0]) | |
if !File.exist?(html_file) || !File.file?(html_file) | |
puts "File #{html_file} does not seem to exist. Try again." | |
exit | |
end | |
begin | |
YotsubatoDownloader.new.download(html_file) | |
rescue Exception => e | |
if e.class != Interrupt | |
puts "Something went wrong!" | |
puts "Error is: #{e.inspect}" | |
end | |
puts "Aborting..." | |
exit | |
end | |
puts "Script finished successfully!" | |
puts "Exiting..." |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment