AI Analyser is a powerful AI-driven tool that enables the analysis of diverse content types - text, audio, images, and video - with ease and precision. Leveraging advanced semantic and sentiment analysis, along with 200+ types of assessments, it evaluates facial expressions, emotional cues, vocal tone, language content, and overall communication quality.

This tool can significantly reduce manual analysis effort - ideal for use cases like job interview evaluations, where AI can help assess honesty, clarity, and positivity across multiple formats.

How to Create an AI Analyser

  1. Navigate to AI Analysers:
    Go to the "AI Analysers" section under the "Content" menu and click on "Add New".

  2. Optional Prompt:
    Begin by entering an AI prompt that defines the purpose of the analysis (e.g., “I want to analyze recorded job interviews and assess honesty, clarity, and positive attitude.”). This step is optional but can improve analysis accuracy.

  3. Fill in the Analyser Form:

    • Name: Provide a name for the analyser.

    • Description: Briefly describe its purpose or focus.

    • Emotion Parameters: Select the emotions you want the analyser to detect and report.

    • Analyser Type:

      • Static: Admin uploads files manually.

      • Dynamic: Users upload files through a task-based module.

    • Input Type: Choose the content type to be analyzed - Text, Image, Audio, or Video.

  4. Save the Form:
    Once all fields are set, save the analyser configuration.

Static Analysis Flow

  1. Upload Content:
    In the analyser grid, click the plus (+) icon under the Actions column to go to the content analyser page.

  2. Submit for Analysis:
    Upload the appropriate files based on the selected input type and click "Submit for Analysis".

  3. Processing Time:
    Analysis duration depends on file size. Check the file's status in the results table at the bottom of the page.

  4. View Results:
    Once analysis is complete, click the eye icon under the Actions column to view the detailed report.

Analysis Report & Metrics

The report provides rich insights, including:

  • Uploaded File Data: Direct reference to the content being analyzed.

  • Key Metrics: Emotional percentage, sentiment score, toxicity indicators, and more.

Based on file type:

  • Text Files:
    Metrics are visually overlaid on the text using color highlights.

  • Images:
    Emotional percentages are shown for each recognized face/expression.

  • Audio:

    • Transcribed text with emotional/sentiment analysis.

    • Graphs for vocal bursts, prosody, and emotional tone for each detected speaker.

  • Video:

    • With audio: Combines speaker-based audio analysis with facial recognition.

    • Without audio: Visual graphs showing emotional expressions for each recognized face.