from pydub import AudioSegment
from pydub.utils import make_chunks
import csv
import os
from pprint import pprint


chunk_s = 2000

natives = {}
with open('speakers_all.csv', 'r') as f:
	reader = csv.reader(f)
	next(reader) # skipting header

	for row in reader:
		native = row[4]
		filename = row[3]

		fn = 'recordings/' + filename + '.wav'
		if native in natives:
			natives[native].append(fn)
		else:
			natives[native] = [fn]

# pprint(natives)

for k, v in natives.items():
	fp = 'recordings/' + k
	os.mkdir(fp)

	for fn in v:
		audio = AudioSegment.from_wav(fn)
		chunks = make_chunks(audio, chunk_s)

		l = len(chunks)
		for i, ch in enumerate(chunks):
			if i == 0 or i == (l - 1):
				continue

			ch.export(fp + '/' + k + str(i) + '.wav', format='wav')