Update app.py

app.py

@@ -42,7 +42,7 @@ def speech_to_speech_translation(audio):
     synthesised_speech = synthesise(translated_text)
     synthesised_speech = (
         synthesised_speech.numpy() * 32767).astype(np.int16)
-    return "", synthesised_speech
+    return [16000, synthesised_speech]
 
 def predict(transType, language, audio, audio_mic = None):
         print("debug1:", audio,"debug2", audio_mic)
@@ -52,7 +52,7 @@ def predict(transType, language, audio, audio_mic = None):
         if transType == "Text":
             return translate(audio), None
         if transType == "Audio":
-            return speech_to_speech_translation(audio)
+            return "",speech_to_speech_translation(audio)
             
 # Define the title etc
 title = "Swedish STSOT (Speech To Speech Or Text)"
@@ -62,11 +62,10 @@ description="Use Whisper pretrained model to convert swedish audio to english (t
 supportLangs = ["Swedish", "French (in training)"]
 transTypes = ["Text", "Audio"]
 
-# examples = [
-#     ["Text", "Swedish", "ex1.mp3", None],
-#     ["Audio", "Swedish", "ex2.mp3", None]
-# ]
-examples = []
+examples = [
+    ["Text", "Swedish", "./ex1.wav", None],
+    ["Audio", "Swedish", "./ex2.wav", None]
+]
 demo = gr.Interface(
     fn=predict,
     inputs=[