Translate speech to any language (Google supported) with Python and Google Translate API

Shelwyn Corte - Nov 8 - - Dev Community

Image description
In this article, we are going to create a speech translator with python using the Google translate API

Installation (Linux):
— pip install SpeechRecognition
— pip install googletrans
— pip install gTTS
— pip install playsound

Installation (Windows):
— pip install SpeechRecognition
— pip install gTTS
— pip install pipwin
— pipwin install pyaudio
— pip install playsound==1.2.2
— pip install googletrans==4.0.0-rc1

Lets import the required modules

import speech_recognition as sr
from googletrans import Translator
from gtts import gTTS
from playsound import playsound
Enter fullscreen mode Exit fullscreen mode

Create an object of the translator class

translator = Translator()
Enter fullscreen mode Exit fullscreen mode

We will now use the default microphone as the audio source, listen to the phrase and extract it into audio data

r = sr.Recognizer()
with sr.Microphone() as source:
    print("Speak Now:")
    audio = r.listen(source)
Enter fullscreen mode Exit fullscreen mode

Set the destination language, you can get a list of all language codes here [https://meta.wikimedia.org/wiki/Template:List_of_language_names_ordered_by_code]

language_to_translate='en'
Enter fullscreen mode Exit fullscreen mode

The below section will print the recognized speech, set the language to be translated to and use the google API to to translate the recognized speech. We will also print the detected text and the translated text on the console

print("Recognized as: ", r.recognize_google(audio))
language = language_to_translate
translations = translator.translate(r.recognize_google(audio), dest=language)
print(translations.origin, ' -> ', translations.text)
Enter fullscreen mode Exit fullscreen mode

Finally we will save the translated text as an mp3 audio file using Google Text-to-Speech and then play it using the playsound library

myobj = gTTS(text=translations.text, lang=language)
myobj.save(tr + ".mp3")
playsound(tr + ".mp3")
Enter fullscreen mode Exit fullscreen mode

*Complete Code:
*

import speech_recognition as sr
from googletrans import Translator
from gtts import gTTS
from playsound import playsound

translator = Translator()
r = sr.Recognizer()
with sr.Microphone() as source:
    print("Speak Now:")
    audio = r.listen(source)

language_to_translate='en'
try:
    print("Recognized as: ", r.recognize_google(audio))
    language = language_to_translate
    translations = translator.translate(r.recognize_google(audio), dest=language)
    print(translations.origin, ' -> ', translations.text)
    myobj = gTTS(text=translations.text, lang=language)
    myobj.save(tr + ".mp3")
    playsound(tr + ".mp3")
except Exception as e:
    print(e)
Enter fullscreen mode Exit fullscreen mode
. . .