update desc

app.py

@@ -117,10 +117,16 @@ if __name__ == "__main__":
         )
         gr.HTML(
             f"""
-            Speed comparison between <a href="https://huggingface.co/openai/whisper-large-v2"> Whisper</a> 
-            and <a href="https://huggingface.co/distil-whisper/distil-large-v2"> Distil-Whisper</a>. Both models use the <a href="https://huggingface.co/distil-whisper/distil-large-v2#long-form-transcription"> chunked long-form transcription algorithm</a> 
-            in 🤗 Transformers with Flash Attention support. To ensure fair usage of the Space, we ask that audio 
-            file inputs are kept to < 30 mins.
+            <a href="https://huggingface.co/distil-whisper/distil-large-v2"> Distil-Whisper</a> is a distilled variant 
+            of the <a href="https://huggingface.co/openai/whisper-large-v2"> Whisper</a> model by OpenAI. Compared to Whisper, 
+            Distil-Whisper runs 6x faster with 50% fewer parameters, while performing to within 1% word error rate (WER) on
+            out-of-distribution evaluation data.
+            
+            In this demo, we perform a speed comparison between Whisper and Distil-Whisper in order to test this claim.
+            Both models use the <a href="https://huggingface.co/distil-whisper/distil-large-v2#long-form-transcription"> chunked long-form transcription algorithm</a> 
+            in 🤗 Transformers, as well as Flash Attention. To use Distil-Whisper yourself, check the code examples on the
+            <a href="https://github.com/huggingface/distil-whisper#1-usage"> Distil-Whisper repository</a>. To ensure fair 
+            usage of the Space, we ask that audio file inputs are kept to < 30 mins.
             """
         )
         audio = gr.components.Audio(type="filepath", label="Audio input")