Speech_Recognizance_Google_API

Windows application with Python using natural language processing and speech recognition libraries as well as libraries for audio processing.

The main goal of the application is to transcribe audio recordings to text and to analyze the generated text using natural language processing and search for given keywords as well as identify common patterns.

---

Step by step install instructions for dependencies:

####Speech Recognition

pip install SpeechRecognition

####Pydub

pip install pydub

####Nltk

pip install nltk install

####PyAudio Download corresponding PyAudio wheel from https://www.lfd.uci.edu/~gohlke/pythonlibs/

pip install C:/full-package-path/package-name.whl

####Stanza

pip install stanza

>>> import stanza
>>> stanza.download('pt')

---

###Creating the executable

After downloading the portuguese model, create a folder called Stanza in the same directory as Main.py.

Place pt folder and resources.json file inside Stanza folder

####Ffmpeg Download Ffmpeg binary from https://ffmpeg.org/download.html.

Place the Ffmpeg folder (with the ffmpeg.exe and ffprobe.exe files) inside the project directory, at the same level as Main.py

####pyinstaller

pip install pyinstaller

pyinstaller --add-data "Ffmpeg/bin/ffmpeg.exe;." --add-data "Ffmpeg/bin/ffprobe.exe;." --add-data "Stanza;Stanza" --hidden-import=stanza Main.py

###Pyinsaller instructions breakdown

--add-data "Ffmpeg/bin/ffmpeg.exe;."

Will add ffmpeg.exe to binary folder after it is generated. Same with ffprobe.exe

--add-data "Stanza;Stanza"

Will add contents of Stanza folder (that should be located in the same directory as Main.py and that should contain pt folder and resources.json file) to binary folder after it is generated.

--hidden-import=stanza

Will add stanza module as a hidden import