It was really easy to introduce. homebrew Thank you.
$ brew install espeak
$ espeak "hello world"
$ brew install open-jtalk
This alone Dictionary files (/usr/local/Cellar/open-jtalk/1.10_1/dic/) The audio file (/usr/local/Cellar/open-jtalk/1.10_1/voice/) is also included.
Female voice: mei_normal.htsvoice
$ echo "I am a cat. There is no name yet. I have no idea where I was born. I remember only crying in a dim and damp place. I saw human beings for the first time here." > sample.txt
$ open_jtalk \
-x /usr/local/Cellar/open-jtalk/1.10_1/dic \
-m /usr/local/Cellar/open-jtalk/1.10_1/voice/mei/mei_normal.htsvoice \
-ow out.wav sample.txt
$ afplay out.wav
Male voice: nitech_jp_atr503_m001.htsvoice
$ open_jtalk \
-x /usr/local/Cellar/open-jtalk/1.10_1/dic \
-m /usr/local/Cellar/open-jtalk/1.10_1/voice/m100/nitech_jp_atr503_m001.htsvoice \
-ow out.wav sample.txt
$ afplay out.wav
Run from python. Create and save the following jtalk.py
.
jtalk.py
# coding: utf-8
import subprocess
from datetime import datetime
def jtalk(t):
open_jtalk = ['open_jtalk']
mech = ['-x', '/usr/local/Cellar/open-jtalk/1.10_1/dic']
htsvoice = ['-m', '/usr/local/Cellar/open-jtalk/1.10_1/voice/mei/mei_normal.htsvoice']
speed = ['-r', '1.0']
outwav = ['-ow', 'out.wav']
cmd = open_jtalk + mech + htsvoice + speed + outwav
c = subprocess.Popen(cmd, stdin=subprocess.PIPE)
c.stdin.write(t)
c.stdin.close()
c.wait()
aplay = ['afplay', 'out.wav']
wr = subprocess.Popen(aplay)
def say_datetime():
d = datetime.now()
text = '%s month%s day,%s time%s minutes%s seconds' % (d.month, d.day, d.hour, d.minute, d.second)
jtalk(text)
if __name__ == '__main__':
say_datetime()
$ python jtalk.py
Used as a python module
$ python
>>> import jtalk
>>> jtalk.jtalk ("Rosei's Richou was joined by Gakusai Saiei, Tenpo Matsunen, and Kobo at a young age, and was then supplemented by Kounani.")
Reference: http://qiita.com/kkoba84/items/b828229c374a249965a9
Recommended Posts