How to save the output of PyTTSx to wav file

Question

I'm trying to find a solution why my code doesn't work properly. I used solution from Recording synthesized text-to-speech to a file in Python, and it kinda didnt worked out for me. The question is why 2 methods/functions text_to_wav and all_texts_to_files doesnt work for me.

import json
import pyttsx
from openpyxl import load_workbook
import subprocess

class Ver2ProjectWithTTS(object):

    def __init__(self):
        self.list_merge = []

    def do_the_job(self):
        self.read_json_file()
        self.read_xml_file()
        #self.say_something()
        self.all_texts_to_files()

    def read_json_file(self):
        with open("json-example.json", 'r') as df:
            json_data = json.load(df)
            df.close()
        for k in json_data['sentences']:
            text_json = k['text']
            speed_json = int(k['speed'])
            volume_json = float(k['volume'])
            dict_json = {'text': text_json, 'speed': speed_json, 'volume': volume_json}
            self.list_merge.append(dict_json)

    def read_xml_file(self):
        tree = et.parse('xml-example.xml')
        root = tree.getroot()
        for k in range(0, len(root)):
            text_xml = root[k][0].text
            speed_xml = int(root[k][1].text)
            volume_xml = float(root[k][2].text)
            dict_xml = {'text': text_xml, 'speed': speed_xml, 'volume': volume_xml}
            self.list_merge.append(dict_xml)

    def say_something(self):
        for item in self.list_merge:
            engine = pyttsx.init()
            engine.getProperty('rate')
            engine.getProperty('volume')
            engine.setProperty('rate', item['speed'])
            engine.setProperty('volume', item['volume'])
            engine.say(cleared_text)
            engine.runAndWait()

    def text_to_wav(self, text, file_name):
        subprocess.call(["espeak", "-w"+file_name+".wav", text])

    def all_texts_to_files(self):
        for item in self.list_merge:
            cleared_text = self.clear_text_from_underscores(item['text'])
            self.text_to_wav(cleared_text, item['text'])

if __name__ == '__main__':
    a = Ver2ProjectWithTTS()
    a.do_the_job()

Error code here:

#In my project:
line 91, in <module> a.do_the_job()
line 21, in do_the_job self.all_texts_to_files()
line 85, in all_texts_to_files self.text_to_wav(cleared_text, item['text'])
line 80, in text_to_wav subprocess.call(["espeak", "-w"+file_name+".wav", text])
#in subprocess:
line 523, in call return Popen(*popenargs, **kwargs).wait()
line 711, in __init__ errread, errwrite)
line 959, in _execute_child startupinfo)
WindowsError: [Error 2] The system cannot find the file specified

Answer 1

from gtts import gTTS
import os
tts = gTTS(text='hi how r u', lang='en')
tts.save("good.wav")
os.system("mpg321 good.wav")

this code will output will be saved in ur python folder where u installed. for various audio format just change extension file.

Answer 2

Assuming you are using Python on a win os, you will need to specify the full path to the subprocess, and of course the full output file path eg;

espeak_path = "C:/Program Files/eSpeak/command_line/espeak.exe"
file_name = "C:/temp/test"
subprocess.call([espeak_path,"-w"+file_name+".wav", text])

Answer 3

Just in case somebody is wondering how to do this on linux instead of windows:

os.system("espeak \"example text to speech\" --stdout > myfile.wav")