Update app.py

app.py

@@ -5,12 +5,12 @@ import torch
 import gradio as gr
 from transformers import Wav2Vec2Tokenizer, Wav2Vec2ForCTC
 nltk.download("punkt")
+
 #Loading the pre-trained model and the tokenizer
 model_name = "facebook/wav2vec2-base-960h"
 tokenizer = Wav2Vec2Tokenizer.from_pretrained(model_name)
 model = Wav2Vec2ForCTC.from_pretrained(model_name)
 def load_data(input_file):
-
   #reading the file
   speech, sample_rate = librosa.load(input_file)
   #make it 1-D
@@ -23,7 +23,8 @@ def load_data(input_file):
   def correct_casing(input_sentence):
     sentences = nltk.sent_tokenize(input_sentence)
   return (' '.join([s.replace(s[0],s[0].capitalize(),1) for s in sentences]))
-  def asr_transcript(input_file):
+
+def asr_transcript(input_file):
     speech = load_data(input_file)
   #Tokenize
     input_values = tokenizer(speech, return_tensors="pt").input_values
@@ -37,9 +38,9 @@ def load_data(input_file):
     transcription = correct_casing(transcription.lower())
   return transcription
 
-    gr.Interface(asr_transcript,
-             inputs = gr.inputs.Audio(source="microphone", type="filepath", optional=True, label="Speaker"),
-             outputs = gr.outputs.Textbox(label="Output Text"),
-             title="ASR using Wav2Vec 2.0",
-             description = "This application displays transcribed text for given audio input",
-             examples = [["Test_File1.wav"], ["Test_File2.wav"], ["Test_File3.wav"]], theme="grass").launch()
+gr.Interface(asr_transcript,
+inputs = gr.inputs.Audio(source="microphone", type="filepath", optional=True, label="Speaker"),
+outputs = gr.outputs.Textbox(label="Output Text"),
+title="ASR using Wav2Vec 2.0",
+description = "This application displays transcribed text for given audio input",
+examples = [["Test_File1.wav"], ["Test_File2.wav"], ["Test_File3.wav"]], theme="grass").launch()