Create a new Python Virtual Environment:
conda create -n rasa281 python=3.8
Activate the environment
conda activate rasa281
Install the Specified pip
version
pip install pip==20.2
Needs Microsoft Build Tool already installed as a dependency in RASA
Install the RASA Specified version
pip install rasa==2.8.1
Install the rasa-sdk specified version
pip install rasa-sdk==2.8.1
So far, We you have installed the followings:
Rasa Version : 2.8.1
Minimum Compatible Version: 2.8.0
Rasa SDK Version : 2.8.6
Rasa X Version : None
Python Version : 3.8.13
Operating System : Windows-10-10.0.19044-SP0
Python Path : C:\Users\username\.conda\envs\rasa281\python.exe
You can get above with this command:
rasa --version
Now, You can create an initial project of RASA and then further provide directory path
rasa init
For Installation of RASA-X first install some dependencey libraries. (press y when prompt for options)
pip install SQLAlchemy==1.3.22
conda install ujson
conda install tensorflow
pip install rasa-x==0.39.3 --extra-index-url https://pypi.rasa.com/simple
Now, Check for whether RASA-X has been installed
rasa --version
Rasa Version : 2.8.1
Minimum Compatible Version: 2.8.0
Rasa SDK Version : 2.8.6
Rasa X Version : 0.39.3
Python Version : 3.8.13
Operating System : Windows-10-10.0.19044-SP0
Python Path : C:\Users\username\.conda\envs\rasa281\python.exe
rasa 2.8.1 with Python virtual environment 3.8 has been installed
[For Running on WhatsApp relevent commands]
$ rasa run -m models --enable-api --cors "*"
Or we can also use to run rasa server:
$ rasa run --log-file out.log --cors * --enable-api
$ ngrok http 5005
For voice chatbot Libraries
For Automatic Speech Recognition (ASR) | Speech To Text (STT)
pip install SpeechRecognition PyAudio
For Text to Speech (TTS)
pip install gtts
[For Windows]
pip install mpyg321
[For Ubuntu]
sudo apt-get install mpg321
For Running Endpoint:
rasa run -m models --endpoints endpoints.yml --port 5002 --credentials credentials.yml