Update README.md

README.md

@@ -10,5 +10,34 @@ pinned: false
 license: apache-2.0
 short_description: an emotional recognition model
 ---
+# Audio Emotion Analyzer with Gemma
+
+This application uses Google's Gemma model to analyze emotions from audio files through acoustic feature analysis.
+
+## Features
+
+- 🎵 Upload audio files or record directly
+- 🎭 Emotion detection using acoustic features
+- 🤖 Powered by Google's Gemma model
+- 📊 Detailed analysis with confidence scores
+- 🎨 Beautiful visual interface
+
+## How it works
+
+1. Extracts acoustic features from audio (MFCC, spectral centroid, pitch, energy, etc.)
+2. Converts features to descriptive text
+3. Uses Gemma to analyze emotional content based on acoustic characteristics
+4. Provides structured emotion analysis with reasoning
+
+## Supported emotions
+
+- Happy, Sad, Angry, Fearful, Disgusted
+- Surprised, Neutral, Excited, Calm, Anxious
+
+## Usage
+
+1. Upload an audio file or record your voice
+2. Wait for the AI analysis
+3. View the emotional analysis with confidence scores and reasoning
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference