Update usage.md

usage.md

@@ -1,9 +1,20 @@
-**Usage:**
 
-This demo takes a audio sample and enhances it using DeepFilterNet2.
-Upload a (noisy) speech sample. You may optionally add some additional background noise to the input sample.
-If no samples are provided, a default will be used.
+### 🎧 DeepFilterNet Voice Denoiser (Extended)
 
-Long audio files will be trimmed to 10s.
+This demo enhances voice recordings using **DeepFilterNet2**, a powerful AI model for **real-time speech denoising**.
 
-DeepFilterNet2 [(link)](https://github.com/Rikorose/DeepFilterNet) is used to denoise the noisy mixture.
+#### 🔹 How to Use
+
+1. **Upload** your noisy audio (`.wav` and `.mp3`) — up to **1 hour long**.
+2. *(Optional)* Add background noise to test the filter.
+3. Click **Submit** to process and compare the **original** and **denoised** results.
+
+#### 🔹 What It Does
+
+Removes unwanted background sounds like fan noise, chatter, hums, or ambient noise — producing a clearer, more natural voice.
+
+#### 🔹 About
+
+Built from [Rikorose/DeepFilterNet](https://github.com/Rikorose/DeepFilterNet) and expanded to handle **long-form audio** (podcasts, interviews, lectures).
+
+---