Update app.py

app.py

@@ -11,15 +11,7 @@ def convert(inputfile, outfile):
         file_type="wav", channels=1, encoding="signed-integer", rate=16000, bits=16
     )
     sox_tfm.build(inputfile, outfile)
-def parse_transcription(wav_file):
-    filename = wav_file.name.split('.')[0]
-    convert(wav_file.name, filename + "16k.wav")
-    speech, _ = sf.read(filename + "16k.wav")
-    input_values = processor(speech, sampling_rate=16_000, return_tensors="pt").input_values
-    logits = model(input_values).logits
-    predicted_ids = torch.argmax(logits, dim=-1)
-    transcription = processor.decode(predicted_ids[0], skip_special_tokens=True)
-    return transcription
+    
     
 
 model_translate = M2M100ForConditionalGeneration.from_pretrained("facebook/m2m100_418M")
@@ -37,6 +29,18 @@ def translate(text):
 processor = Wav2Vec2Processor.from_pretrained("Harveenchadha/vakyansh-wav2vec2-hindi-him-4200")
 model = Wav2Vec2ForCTC.from_pretrained("Harveenchadha/vakyansh-wav2vec2-hindi-him-4200")
 
+def parse_transcription(wav_file):
+    filename = wav_file.name.split('.')[0]
+    convert(wav_file.name, filename + "16k.wav")
+    speech, _ = sf.read(filename + "16k.wav")
+    input_values = processor(speech, sampling_rate=16_000, return_tensors="pt").input_values
+    logits = model(input_values).logits
+    predicted_ids = torch.argmax(logits, dim=-1)
+    transcription = processor.decode(predicted_ids[0], skip_special_tokens=True)
+    return transcription, translate(transcription)
+
+
+
 output1 = gr.outputs.Textbox(label="Hindi Output from ASR")
 output2 = gr.outputs.Textbox(label="English Translated Output")