I'm learning to mess with Pyqt. I'm putting together a simple application, where there will be a voice notification when a new request is available.
I would like my application written in Pyqt4 to "transform" a text into a voice.
I was able to do this with HTML5 and Javascript, but now I would like to know if there is any way to do something like Pyqt.
Is there any functionality in Pyqt or that can be used together with Pyqt to do Text to Speech (text to speech)?
I believe that PyQt4 does not support this, and pyqt5 may only come in support in the future (apparently QtSpeech is very recent), so using PyQt is not an option, however I found this: link (it's a bit outdate 2012), but it's cross-platform and support Python2 and this link
Drivers required per environment:
nsss - NSSpeechSynthesizer in Mac OS X 10.5 and higher
Details: link
sapi5 - SAPI5 on Windows
Download at link
espeak - eSpeak in distros linux
Install on Debian and Ubuntu:
sudo apt-get install espeak
To install manually: link
Then you can install it via PIP:
pip install pyttsx
See an example:
import pyttsx
engine = pyttsx.init()
engine.say('Hello World!')
engine.runAndWait()
Links:
In the releases of github link and maybe in pip only version 1.1 appears, but in the doc already informs about the version 1.2, in the case it would just install this link manually
The lib by itself does not have the languages installed, who has to have installed is the driver used, in the case of eSpeak I think it already comes with Portuguese, to configure the language in the lib one can use something like, the for is to get all the voices:
engine = pyttsx.init()
voices = engine.getProperty('voices')
for voice in voices:
engine.setProperty('voice', voice.id) # troca a voz
engine.say('The quick brown fox jumped over the lazy dog.')
engine.runAndWait()
Then you will select the voice you want, note that voices = engine.getProperty('voices') returns the class pyttsx.voice.Voice class, which has the following properties:
.age returns age in years as integer
.gender returns a String that contains the voice genre being female , male or neutral
.id is the voice identifier to use in pyttsx.engine.Engine.setPropertyValue()
.languages returns a list of strings that shows the languages supported by the voice .name returns a String containing a "human" name to help assimilate language, gender, etc. In case you can use voice.languages to get the Portuguese language ID and so set it.
If there is no Portuguese language in the list it is because the driver does not support or does not have the language installed, in the case of Linux e eSpeak already comes with several languages:
Windows also seems to already have some including the ones in Portuguese:
What I liked about Espeek is that it has support for Windows and MacOSX, I just do not know if lib is compatible.
gTTS (Google Text to Speech) is an interface that transforms text into audio and saves it to an .mp3
To install run PIP:
pip install gTTS
Usage:
from gtts import gTTS
tts = gTTS(text='Hello', lang='en')
tts.save("hello.mp3") #Salva o arquivo
In Portuguese:
from gtts import gTTS
tts = gTTS(text='Olá', lang='pt-br')
tts.save("ola.mp3") #Salva o arquivo
Supported Languages: link