Project Prerequisites<\/h3>\n
It is suggested to have prior knowledge in Python and basic ideas about the Tkinter module. All the above modules can be installed using the following commands.<\/p>\n
pip install tkinter\r\npip install pyttsx3\r\npip install pydub\r\npip install SpeechRecognition\r\n<\/pre>\nProject Structure<\/h3>\n
Steps to build the project are:<\/p>\n
1. Importing the required modules
\n2. Creating the main window
\n3. Function to convert text to audio
\n4. Writing the function to create window to take text input
\n5. Function to create a window for showing text output<\/p>\n6. Writing function to get the audio<\/p>\n
1. Importing the required modules<\/h4>\n
import tkinter \r\nfrom tkinter import filedialog\r\nfrom tkinter import *\r\n\r\nfrom path import Path\r\nimport pyttsx3\r\nfrom speech_recognition import Recognizer, AudioFile\r\nimport speech_recognition as sr\r\nfrom pydub import AudioSegment\r\nimport os\r\nfrom time import sleep\r\n<\/pre>\nCode explanation:<\/strong><\/p>\n
Here, we first initially start by importing all the required modules discussed above.<\/p>\n
2. Creating the main window<\/h4>\n
wn = tkinter.Tk() \r\nwn.title(\"PythonGeeks Text to Audio and Audio to Text converter\")\r\nwn.geometry('700x300')\r\nwn.config(bg='LightBlue1')\r\n \r\nLabel(wn, text='TechVidvan Text to Audio and Audio to Text converter',\r\n fg='black', font=('Courier', 15)).place(x=40, y=10)\r\n\r\nglobal textBox,showText,command\r\ngo=1\r\n\r\nButton(wn, text=\"Convert Text to Audio\", bg='ivory3',font=('Courier', 15),\r\n command=text_to_audio).place(x=230, y=80)\r\nButton(wn, text=\"Convert Audio to Text\", bg='ivory3',font=('Courier', 15),\r\n command=audio_to_text).place(x=230, y=150)\r\n\r\nwn.mainloop()\r\n<\/pre>\nCode explanation:<\/strong>
\nIn this step we create a window \u2018wn\u2019 using Tkinter to create a window with two buttons for converting text to audio and audio to text vice versa. On clicking this button, respective functions text_to_audio() or audio_to_text(), give as command parameters, get executed.<\/p>\n3. Function to convert text to audio<\/h4>\n
#text to voice\r\nvoiceEngine = pyttsx3.init('sapi5')\r\nvoices = voiceEngine.getProperty('voices')\r\nvoiceEngine.setProperty('voice', voices[1].id)\r\n\r\ndef speak():\r\n global textBox\r\n text=textBox.get(1.0, \"end-1c\")\r\n print(text)\r\n voiceEngine.say(text)\r\n voiceEngine.runAndWait()\r\n<\/pre>\nCode explanation:<\/strong>
\nThis function speak() takes text as input and converts it into audio output by creating a voice engine using the pyttsx3 module. Here, the getProperty() and setProperty() methods are used to access the device audio system.<\/p>\nAnd the text to be spoken is taken from the widget \u2018textBox\u2019 using the get() method. Finally, the say() method is the function that does this job of giving audio output.<\/p>\n
4. Writing the function to create window to take text input<\/h4>\n
def text_to_audio():\r\n #Creating a window \r\n global textBox\r\n wn1 = tkinter.Tk() \r\n wn1.title(\"TechVidvan Text to Audio converter\")\r\n wn1.geometry('500x500')\r\n wn1.config(bg='snow3')\r\n \r\n Label(wn1, text='TechVidvan Text to Audio converter',\r\n fg='black', font=('Courier', 15)).place(x=60, y=10)\r\n \r\n v=Scrollbar(wn1, orient='vertical')\r\n v.pack(side=RIGHT, fill='y')\r\n textBox=Text(wn1, font=(\"Calibre, 14\"), yscrollcommand=v.set)\r\n textBox.focus()\r\n textBox.place(x=20, y=80,width=450,height=300)\r\n \r\n v.config(command=textBox.yview)\r\n Button(wn1, text=\"Convert\", bg='ivory3',font=('Courier', 13),\r\n command=speak).place(x=230, y=400)\r\n \r\n wn1.mainloop()\r\n<\/pre>\nCode explanation:<\/strong>
\nThis function runs when the user clicks the \u2018Convert Text to Audio\u2019 button on the main window. This function contains a scrollable text box where the user can give the input text that is to be converted. On writing text and clicking the \u2018Convert\u2019 button, the function speak() executes.<\/p>\n5. Function to create a window for showing text output<\/h4>\n
def audio_to_text():\r\n #Creating a window\r\n global showText\r\n wn2= tkinter.Tk() \r\n wn2.title(\"TechVidvan Audio to Text converter\")\r\n wn2.geometry('500x500')\r\n wn2.config(bg='snow3')\r\n \r\n res=IntVar()\r\n pdfPath = StringVar(wn2) #Variable to get the PDF path input\r\n \r\n Label(wn2, text='TechVidvan Audio to PDF converter',\r\n fg='black', font=('Courier', 15)).place(x=60, y=10)\r\n\r\n #Getting the PDF path input\r\n Label(wn2, text='Click the start and end buttons to speak and end speech').place(x=20, y=50)\r\n \r\n Button(wn2, text='Start', bg='ivory3',font=('Courier', 13),\r\n command=takeCommand).place(x=100, y=100)\r\n\r\n #Button to select the audio file and do the conversion \r\n Button(wn2, text='Stop', bg='ivory3',font=('Courier', 13),\r\n command=stop).place(x=200, y=100)\r\n \r\n v=Scrollbar(wn2, orient='vertical')\r\n v.pack(side=RIGHT, fill='y')\r\n showText=Text(wn2, font=(\"Calibre, 14\"), yscrollcommand=v.set)\r\n showText.focus()\r\n showText.place(x=20, y=130,width=450,height=300)\r\n \r\n v.config(command=showText.yview)\r\n wn2.mainloop() #Runs the window till it is closed\r\n<\/pre>\nCode explanation:<\/strong>
\nThis function executes when the user clicks the \u2018Convert Audio to Text\u2019 button on the main window. It has two buttons to start and stop speaking. And everytime the user clicks the stop button the voice gets captured using the takeCommand() function, till the stop button is clicked.<\/p>\nWhen the stop button is clicked, the function stop() gets executed, which converts audio to text from using the below, which is shown on the text box.<\/p>\n
6. Writing function to get the audio<\/h4>\n
def takeCommand():\r\n global showText,go,command\r\n showText.delete(1.0,\"end\")\r\n showText.insert(END,\"Listening....\")\r\n \r\n recog = sr.Recognizer()\r\n command=''\r\n \r\n while go:\r\n with sr.Microphone() as source:\r\n print(\"Listening to the user\")\r\n recog.pause_threshold = 1\r\n userInput = recog.listen(source)\r\n\r\n try:\r\n print(\"Recognizing the command\")\r\n command=command+' '+( recog.recognize_google(userInput, language ='en-in'))\r\n print(f\"Command is: {command}\\n\")\r\n\r\n except Exception as e:\r\n print(e)\r\n print(\"Unable to Recognize the voice.\")\r\n return \"None\"\r\n\r\ndef stop():\r\n global go,command\r\n print(\"q pressed, exiting...\")\r\n go = 0\r\n showText.delete(1.0,\"end\")\r\n showText.insert(END,command)\r\n<\/pre>\nCode explanation:<\/strong>
\nThe takeCommand() shows the \u2018Listening….\u2019 command on the text box indicating that the user’s audio is getting captured. It then creates a Recognizer() object to get the audio input from one of the microphones attached to the device. Here, \u2018go\u2019 variable decides whether the user audio input should be taken or not. It keeps listening to the audio which is converted to texe using the recognize_google() method and stored in \u2018command\u2019 variable.<\/p>\nAnd the stop function stops the process of taking audio and shows the information stored in the \u2018commad\u2019 variable on the text box \u2018showText\u2019.<\/p>\n
Output of Speech to text and text to Speech Python Project<\/h3>\n
Python Text to Speech conversion GUI<\/strong><\/p>\n
![\"python](\"https:\/\/techvidvan.com\/tutorials\/wp-content\/uploads\/sites\/2\/2022\/06\/python-text-to-speech-conversion.webp\")
<\/a><\/p>\n![\"python](\"https:\/\/techvidvan.com\/tutorials\/wp-content\/uploads\/sites\/2\/2022\/06\/python-speech-to-text-conversion.webp\")
<\/a><\/p>\n