Upload app.py

app.py

@@ -4,9 +4,6 @@ import whisper
 import validators
 import gradio as gr
 
-import nltk
-nltk.download()
-
 from wordcloud import WordCloud, STOPWORDS
 
 from scipy.io.wavfile import write
@@ -143,9 +140,13 @@ with demo:
     [13 mins speech](https://www.youtube.com/watch?v=nepOSEGHHCQ)
     
     1. Type in a youtube URL or upload an audio file
-    2. Generate transcription with Whisper (Currently English Only)
+    2. Generate transcription with Whisper (English Only)
     3. Summarize the transcribed speech
-    4. A little wordcloud for you as well
+    4. Generate summary's speech with ESPNet model
+    
+    model references:
+    - [Whisper](https://github.com/openai/whisper)
+    - [ESPNet](https://github.com/espnet/espnet_model_zoo)
     """)
 
     # data preparation