Skip to main content

ChatGpt X Oracle Digital Assistant Integration

 Introduction

In this article I will be explaining about integrating the ChatGpt with Oracle Digital Assistant.

NOTE: This is not a suggestion to have ChatGPT in your ODA, this article is just to showcase the flexibility of the ODA to easily integrate with anything.

Main article

ChatGPT

ChatGPT is an AI language model developed by OpenAI, capable of processing natural language input from users and generating coherent and relevant responses. It has been trained on an enormous corpus of text data using unsupervised learning, allowing it to generate responses that are not just grammatically correct, but also contextually appropriate and engaging.

ChatGPT is particularly useful in scenarios where there is a need for rapid and accurate responses to user queries, such as in customer service or personal assistants. However, it may struggle to generate appropriate responses in certain situations and may not be suitable for scenarios requiring a high degree of personalization or customization.

Overall, ChatGPT is a powerful tool that has the potential to revolutionize the way we interact with technology, and we can expect even more advanced language models to emerge in the future.

Oracle Digital Assistant

Oracle Digital Assistant is a platform that allows businesses to develop and deploy conversational interfaces that can interact with customers, employees, and other stakeholders using natural language. It is a cloud-based AI-powered solution that enables businesses to build chatbots, voice assistants, and messaging apps that can automate routine tasks and provide personalized support to users.

Oracle Digital Assistant offers a range of pre-built templates and tools that allow businesses to easily develop and customize their conversational interfaces without requiring extensive coding expertise. The platform integrates with a range of back-end systems, including CRM, ERP, and HR systems, enabling it to access relevant data and provide users with personalized responses.

Oracle Digital Assistant can be used in a range of industries, including healthcare, finance, and retail, to improve customer engagement, streamline operations, and increase productivity.

 

Hence if we bring the AI power of ChatGPT and the custom Digital assistant development power of ODA under one roof would lead to numerous opportunities in any sector.

 

Just follow the below steps to integrate the ODA and ChatGpt:

  1. Go to API services section of ODA

                                                       




















2. Create a new connection with name Chat GPT and provide the below details:



You can provide the sample payload in the POST payload section as below and test it

{

    "model": "text-davinci-003",

    "prompt": "share some fascinating details about india",

    "temperature": 0,

    "max_tokens": 100

}



You can get the bearer token once you create your account in chatGPT go to below URL:

https://platform.openai.com/account/api-keys


Click on the API keys followed by create new secret key.

You will get an auto-generated key that can be used as the bearer token in the connection.

3. Now go to the skills section of ODA


4. Now create a new skill with name “ChatGPT-ODA-Integration"

No need to create any entities or intents, directly navigate to the dialog flow page and click on unresolved intent.

5.  Create a variable using set variable state as shown below:

This basically stores the user input provided

6.  Call the ChatGPT API using the call REST service component.


Provide the request body as below:


7. Now to store the output of the API create a flow level variable as below:



8.Now to display the output of ChatGPT use the send Message component as shown.


Your end flow should look like this


Let’s test the chatbot created so far.

Click on the Preview button on the TOP right corner

Just provide the inputs as if you are interacting with ChatGPT.



As observed from the above the above screen shots the integration is working as expected.

Hope you all liked the article and learnt something new for the day.



-Preetham Konjeti


Labels:

Comments

Post a Comment

Popular posts from this blog

Convert App based Integration to Scheduled Integration without wrapper in OICS

we all know that converting a scheduler integration to REST is quite easy , where as the reverse is not possible and the only way to achieve it is through creating a wrapper scheduler integration and invoking the trigger one as child from there. But there is another way through which we can achieve the direct conversion. For that please refer to below example. Note : Every step mentioned has to be done very carefully Create Scheduler Integration. I’ve created a base one in which it invokes a SOAP service.  I’ve created a small trigger integration which I’m going to change to scheduler without any wrapper. Please follow the below steps to convert rest based to scheduler:     1)    Export the integration.     2) Rename the iar file to zip.     3) Now unzip the file.     4)    Compare the properties files of both scheduler and rest based integration and make the changes to rest based as per the scheduler integration After ...
2 comments
Read more

Backup Integrations of an Instance with a single click using OIC

  Introduction One common mistake that most of OICS developers make is not taking the backup of their integrations religiously and if they want to take backups clicking on the export of the integrations individually could be quite tedious. Main Article In this blog I will be explaining about automating the backup process in OICS. Below is the order of the steps that are to be followed: STEP:1 1) Configure a Rest trigger connection as shown below. Provide the below payload in the request section:     {   "IntegrationName”: "" } STEP:2   Configure the new rest connection as mentioned below. (This basically list downs all the integration details as per the provided name).   Provide the Query parameters as below:   Provide the response payload as below: {   "hasMore" : "false",   "items" : [ {     "id" : "",     "name" : "",     "lockedFla...
Post a Comment
Read more