I just upgraded to the Mac OS Mountain Lion and this might already be common knowledge but I just found out that if you go to System Preferences > Dictation > Text to Speech > Customize you can then scroll down the list and select Japanese. Download the file and then set it to your default. Now if you highlight Japanese text, right click and chose Speech it'll read it for you.

I tried it on the Yoshimoto Banana website and it got 行った wrong but whatever... I don't think I'll ever actually use it but it might help people just starting out learning to read or just be something fun to mess around with for 3 minutes.

http://www.yoshimotobanana.com/diary/

Crispy Wrote:I just upgraded to the Mac OS Mountain Lion and this might already be common knowledge but I just found out that if you go to System Preferences > Dictation > Text to Speech > Customize you can then scroll down the list and select Japanese. Download the file and then set it to your default. Now if you highlight Japanese text, right click and chose Speech it'll read it for you.

I tried it on the Yoshimoto Banana website and it got 行った wrong but whatever... I don't think I'll ever actually use it but it might help people just starting out learning to read or just be something fun to mess around with for 3 minutes.

http://www.yoshimotobanana.com/diary/

Is the TTS called "Kyoko"? If so, than I knew someone who had it on Windows, and...let's just say it's not the most accurate. It might be good if there are no other options though. I wish people could download Google's Japanese TTS. Now that would be awesome...

Haha, yeah that's "her". It's pretty lame overall and better (paid) applications do exist but I'm sure there's someone that might find use in this.

If you don't want to make Kyoko the default voice, you can assign a shortcut to a script that speaks text using it.

Hiragana versions of words are sometimes pronounced incorrecly unless you surround them with punctuation characters:

say -v kyoko 'ぜんじつ、さくじょ、「ぜんじつ」、「さくじょ」'

I've used scripts like this to create sound lessons:

words="さいきどう restart
かんきょうせってい preferences"
IFS=$'\n'
i=1000
for line in $words; do
say "[[slnc 1000]]「${line%% *}」" -v kyoko -o /tmp/$((i++)).aif
say "[[slnc 300]][[volm 0.45]] ${line#* }" -o /tmp/$((i++)).aif
done
/usr/local/bin/sox /tmp/*.aif ~/Desktop/combined.aif
rm /tmp/*.aif