Works locally

app.py

@@ -9,7 +9,7 @@ import streamlit as st
 
 def speech_to_text(uploaded):
     model = whisper.load_model('tiny')
-    result = model.transcribe(uploaded)
+    result = model.transcribe(uploaded,verbose=True)
     return f'You said: {result["text"]}'
 
 def segment(nu_speakers):
@@ -28,15 +28,14 @@ def segment(nu_speakers):
 
 def audio_to_df(uploaded):
     monotize(uploaded)
-    model = whisper.load_model('base')
+    model = whisper.load_model('tiny')
     result = model.transcribe('mono.wav',verbose=True,
                           without_timestamps=False)
     tdf = pd.DataFrame(result['segments'])
     return tdf
 
 def monotize(uploaded):
-    print(uploaded.name)
-    cmd = f"ffmpeg -y -i {uploaded.name} -acodec pcm_s16le -ar 16000 -ac 1 mono.wav"
+    cmd = f"ffmpeg -y -i {uploaded} -acodec pcm_s16le -ar 16000 -ac 1 mono.wav"
     subprocess.Popen(cmd, shell=True).wait()
 
 def add_preface(row):
@@ -70,6 +69,7 @@ def transcribe(uploaded, nu_speakers):
 
     lines = []
     for row in binned_df['output'].values:
+        st.write(row)
         lines.append(row)
 
     return '\n'.join(lines)
@@ -95,5 +95,9 @@ if submit:
     bytes_data = uploaded.getvalue()
     with open('temp_audio', 'wb') as outfile:
         outfile.write(bytes_data)
-    speech_to_text('temp_audio')
+    #st.write('Converting audio file.')
+    #monotize('temp_audio')
+    text = transcribe('temp_audio', nu_speakers)
+    
+
     # To read file as bytes: