fixed bug

app.py

@@ -43,7 +43,7 @@ def generate_image(text):
     cond_scale = 3.0
  
     encoded_images = model.generate(
-        tokenized_prompt,
+        **tokenized_prompt,
         random.randint(0, 1e7),
         model.params,
         gen_top_k,
@@ -76,9 +76,13 @@ def parse_transcription(wav_file):
     predicted_ids = torch.argmax(logits, dim=-1)
     transcription = asr_processor.decode(predicted_ids[0], skip_special_tokens=True)
     
+    print(transcription)
+
     # translate to english
     translated = he_en_translator(transcription)[0]['translation_text']
 
+    print(translated) 
+
     # generate image
     image = generate_image(translated)
     return image