Skip to main content

Project Administration

Project administration in Ayushma empowers authorized personnel (admins) with the ability to create, configure, and manage projects that serve as the foundation for AI-assisted medical conversations.

Creating a New Project​

  1. Navigate to Projects: From the admin dashboard, access the "Projects" section.
  2. Initiate Project Creation: Click on the "New Project" button or a similar option to start the creation process.
  3. Provide Project Details: Fill in the essential project information:
    • Title: A concise and descriptive name for the project.
    • Description (Optional): An overview of the project's purpose, scope, and intended use.
  4. Configure AI Settings:
    • Model: Select the desired AI model for the project. Ayushma offers various models, each with its strengths and capabilities. Options may include GPT-3.5, GPT-4, and others.
    • Prompt (Optional): Provide a prompt or set of instructions to guide the AI assistant's behavior and response style within the project's context.
  5. Preset Questions (Optional): Add a list of preset questions relevant to the project's domain. These questions will appear as suggestions for users to initiate conversations.
  6. Save Project: Click the "Create" or "Save" button to finalize the project creation.

Configuring Project Settings​

  • Title and Description: Admins can modify the project's title and description as needed.
  • AI Model: The selected AI model can be changed to adjust the AI assistant's capabilities and performance.
  • Prompt: Admins can refine the prompt to fine-tune the AI's behavior and response style.
  • Preset Questions: The list of preset questions can be updated or expanded to offer users more starting points for conversations.
  • Default Project: Admins can designate a project as the default project. Users will be directed to the default project upon initial login or when no specific project is selected.
  • Archived: Projects can be archived to remove them from active use while retaining their data for future reference or analysis.

Managing Documents​

  1. Access Project Documents: Navigate to the documents section within the selected project.
  2. Add Documents: Click the "Add Document" button or a similar option.
  3. Choose Document Type: Select the type of document you want to add, such as:
    • File: Upload a file from your local system. Supported formats might include PDF, TXT, and others.
    • URL: Provide a URL to a web page containing relevant information.
    • Text: Directly input text content.
  4. Provide Document Details:
    • Title: Enter a title for the document.
    • Description (Optional): Add a description for the document.
  5. Save Document: Click the "Save" button to add the document to the project.

Additional Insights​

  • Speech-to-Text Engine: Admins can choose the speech-to-text engine used for processing voice input within a project. Options may include OpenAI Whisper, Google Speech-to-Text, and potentially self-hosted solutions.
  • Text-to-Speech Engine: Similar to the speech-to-text engine, admins can select the text-to-speech engine for generating audio output. Ayushma may offer options such as OpenAI and Google Text-to-Speech.
  • OpenAI API Key: The code suggests that Ayushma might require an OpenAI API key for certain functionalities, such as accessing advanced AI models or using OpenAI's speech-to-text and text-to-speech services. Admins might need to manage these API keys and ensure their validity.

Project Administration Benefits​

  • Tailored AI Experiences: Project administration allows admins to customize AI assistant experiences for different medical domains or use cases, optimizing performance and relevance for specific contexts.
  • Knowledge Management: Admins can curate and maintain project-specific knowledge bases by adding and managing reference documents, ensuring the AI assistant has access to the most up-to-date and relevant information.
  • Control and Security: Project administration tools enable admins to control access to projects, ensuring data security and privacy.
  • Performance Monitoring: Admins can monitor project usage, analyze conversation history, and gather feedback to continually improve the AI assistant's performance and effectiveness.