Skip to content

Latest commit

 

History

History
97 lines (63 loc) · 6.42 KB

cloud-setup.md

File metadata and controls

97 lines (63 loc) · 6.42 KB
Raw

Cloud Setup

Welcome to the AI Hackathon Starter Kit! These instructions will help you get started with the cloud environment for the hackathon.

The starter kit is designed to be used with the following cloud services:

  1. Azure Speech Services
  2. Azure Text Analytics Services
  3. Open AI Services (You can use either openai.com or the Azure Open AI service. You don't need both.)

Once you have finished reading this guide, go back to the Getting Started page to continue.

Azure Speech Services

Azure Speech Services is a cloud-based API that provides developers with access to speech-to-text and text-to-speech capabilities. The service is available in 3 different tiers: Free, Standard, and Premium. The Free tier is limited to 5,000 transactions per month, while the Standard and Premium tiers are limited to 500,000 transactions per month. For more information on the Azure Speech Services, please visit the Azure Speech Services Documentation.

Create a Speech Service Resource

  1. Navigate to the Azure Portal and login with your Azure account.
  2. Click on the Create a resource button (big plus sign) in the top left corner of the portal.
  3. Search for Speech and select Speech from the search results.
  4. Click on the Create button.
  5. Enter a unique name for your Speech Service resource.
  6. Select the Free pricing tier.
  7. Select the Location closest to you.
  8. Click on the Review + create button.
  9. Click on the Create button.
  10. Once the resource has been created, click on the Go to resource button.
  11. Click on the Keys and Endpoint button in the left navigation bar.
  12. Copy the location & key of the Speech Service resource into a text editor for later. You will use it when setting up your .env file.

Azure Text Analytics Services

Azure Text Analytics Services is a cloud-based API that provides developers with access to sentiment analysis, key phrase extraction, language detection, and entity recognition capabilities. The service is available in 3 different tiers: Free, Standard, and Premium. The Free tier is limited to 5,000 transactions per month, while the Standard and Premium tiers are limited to 500,000 transactions per month. For more information on the Azure Text Analytics Services, please visit the Azure Text Analytics Services Documentation.

Create a Text Analytics Service Resource

  1. Navigate to the Azure Portal and login with your Azure account.
  2. Click on the Create a resource button (big plus sign) in the top left corner of the portal.
  3. Search for Language Service and select Language Service from the search results. (Note: Language Service is different from Language Understanding Service).
  4. Click on the Create button. (Use th default features)
  5. Enter a unique name for your Language Service resource.
  6. Select the Free pricing tier.
  7. Select the Location closest to you.
  8. Click on the Review + create button.
  9. Click on the Create button.
  10. Once the resource has been created, click on the Go to resource button.
  11. Click on the Keys and Endpoint button in the left navigation bar.
  12. Copy the location, key and endpoint of the Language Service resource into a text editor for later. You will use it when setting up your .env file.

Azure Open AI Services

Azure OpenAI Service provides REST API access to OpenAI's powerful language models including the GPT-3, Codex and Embeddings model series. In addition, the new GPT-4 and ChatGPT (gpt-35-turbo) model series are now available in preview. These models can be easily adapted to your specific task including but not limited to content generation, summarization, semantic search, and natural language to code translation. Users can access the service through REST APIs, Python SDK, or web-based interface in the Azure OpenAI Studio.

Create an Open AI Services Resource

Note: Currently, access to this service is granted only by application. You can apply for access to Azure OpenAI by completing the form at https://aka.ms/oai/access. Open an issue on this repo to contact us if you have an issue.

  1. Navigate to the Azure Portal and login with your Azure account.
  2. Click on the Create a resource button (big plus sign) in the top left corner of the portal.
  3. Follow these instructions to create an Azure OpenAI resource.

Create a model in Azure OpenAI Studio

  1. Navigate to the Azure Open AI studio and login with your Azure account.
  2. Choose your Directory, Subscription and Resource where you created the service.
  3. Click on the Deployments navigation item in the left navigation bar.
  4. Click on the Create new deployment button.
  5. Choose the text-davinci-003 model.
  6. Enter a unique name for your model. (Ideally keep it text-davinci-003 so that it's easy to find later.)
  7. Now repeat steps 3-6 for the gpt-35-turbo model.
  8. If you have access to gpt4 model, repeat steps 3-6 for the gpt4 model.
  9. Navigate to the settings cog wheel on the top right(ish) corner of the page.
  10. Select Resource Settings.
  11. Copy the key, endpoint and Location of the Azure Open AI Services resource into a text editor for later. You will use it when setting up your .env file.

Open AI Services (using openai.com)

Note: You don't need both Open AI Services and Azure Open AI Services. You can use either one.

Open AI Services is a cloud-based API that provides developers with access to a variety of AI capabilities. At the tim of writing this document, openai provides an option to "Start for free (Start experimenting with $5 in free credit that can be used during your first 3 months). See Open AI Pricing for more information.

Create an Open AI Services Resource

  1. Navigate to the Open AI Services Portal and login or register.
  2. After setting up your account & billing you will be able to access the API keys.
  3. Copy the key of the Open AI Services resource into a text editor for later. You will use it when setting up your .env file.