I want to use SofTalk from Python

environment

Python 3.8.3
SofTalk 1.93.46
Windows10

Preparation

Download SofTalk Please download the latest version from the following page https://w.atwiki.jp/softalk/pages/15.html

Try using SofTalk from the command line

There is also a way to use Softalk from the command line, as described in the included help.chm. Unzip the DL zip

start ..\softalk\Softalk.exe

This will start SofTalk itself and the following arguments will be available

SofTalk argument quick reference table

Excludes arguments that are not available in slow voice (only available in Aquestalk 10)

type	value	effect	Details	Remarks
/NM:	String	Voice designation	Specify voice by name	Example)/NM:Female 01
/M:	0～?		Specify voice by number
/PR:	String	Preset specification	Specify preset by name	Example)/PR:Initial setting
/N:	0～?		Specify presets by number
/O:	0～300	Interval	声のInterval	If you lower it, the reading speed will only decrease, so there may not be much difference from the speed.
/P:	reading	reading追加	reading(Hiragana),word,all/Distinguish half-width(True/False)	Example)/P:Tesuto,Tｅst,True
			Update previous words for registered words
/Q:	0～5	How to read	0:No hiragana conversion / no intonation	Set for the selected library
			1:No hiragana conversion / with intonation
			2:With hiragana conversion, without intonation
			3:Hiragana conversion Yes / Inflection Yes
			4:With hiragana conversion, without intonation, with mishearing
			5:Hiragana conversion Yes, Inflection Yes, Song Song Yes
/R:		recording	読み上げた文章のrecording	Example)/R:c:\test.wav
/S:	1～300	speed	読み上げのspeed
/T:	7～11	Library specification	Library (engine) specification	0 for slow Reimu, 1 for slow Marisa voice
			7:AquesTalk
			8:SAPI
			9:Speech Platform
			10:AquesTalk2
			11:AquesTalk10
/U:	0～?	Voice designation	Specify voice from library (engine)	Set for the selected library
/V:	0～100	Volume	声のVolume
/W:	Sentence	Sentence	読み上げたいSentence	このオプション以降の文字は全てSentenceとみなす
/X:	0～1	Screen display	0:Display 1:Hide	Valid only at startup, when exiting/close
/Z:	path	XML output	Output specifiable voice in XML format

It doesn't seem to be particularly difficult if you pay attention to the position of / W.

Try running from python

It's a hassle to type from the command line every time Let Python do the hassle

If you want to execute a command from Python, use the subprocess module [^ subprocess]

I didn't know until I wrote this article, but in Python 3.5 and later, you should use ``` subprocess.run () `` instead of subprocess.call () ``.

** The recommended way to start a subprocess is to use the run () function, which can handle all usages. ** For more advanced usage, you can also use the underlying Popen interface directly. ** The run () function was added in Python 3.5 **; If you need to maintain compatibility with previous versions, see the old high-level API section.

Try to write the code

yukkuri ├yukkuritest.py └softalk (decompressed folder) 　└ SofTalk.exe

Imagine a directory structure like this

You can list the commands by space and pass them to the run () function. I thought ... but apparently I have to combine the lists once with a half-width space. So, in simple terms, it looks like this:

`yukkuritest.py`


import os
import subprocess

os.chdir(os.path.dirname(os.path.abspath(__file__)))

_start = "start ..\\yukkuri\\softalk\\SofTalk.exe"
_speed = "/S:120"
_word = "/W:Good morning"

_command = [_start, _speed, _word]

subprocess.run(' '.join(_command), shell=True)

Now you can slowly output your voice from Python

[^ subprocess]: subprocess --- Subprocess management https://docs.python.org/ja/3/library/subprocess.html

Let's make a voice slowly with Python