Update app.py

app.py

@@ -1,53 +1,24 @@
+import gradio as gr
+import numpy as np
+import sounddevice as sd
 from transformers import pipeline
-import soundfile as sf
 
 # Load ASR model
 asr_model = "Abdullah17/whisper-small-urdu"
 asr_pipe = pipeline("automatic-speech-recognition", model=asr_model)
 
-# Rest of your code
-commands = [
-    "نمائندے ایجنٹ نمائندہ",
-    "سم ایکٹیویٹ",
-    "سم بلاک بند",
-    "موبائل پیکیجز انٹرنیٹ پیکیج",
-    "چالان جمع",
-    "گانا سنانا"
-]
+# Function to transcribe the command from audio samples
+def transcribe_the_command(audio_samples):
+    transcript = asr_pipe(np.array(audio_samples))[0]["text"]
+    most_similar_command, reply = find_most_similar_command(transcript, commands)
+    return f"Transcript: {transcript}\nMost Similar Command: {most_similar_command}"
 
-# Function to transcribe the command from audio
-def transcribe_the_command(audio_list):
-    transcriptions = []
-    
-    # Process each audio in the batch
-    for audio_data, sample_rate in audio_list:
-        file_name = "recorded_audio.wav"
-        sf.write(file_name, audio_data, sample_rate)
-        
-        # Convert stereo to mono by averaging the two channels
-        transcript = asr_pipe(file_name)[0]["text"]
-        most_similar_command, reply = find_most_similar_command(transcript, commands)
-        
-        transcriptions.append((transcript, most_similar_command, reply))
-    
-    return transcriptions
-# from transformers import pipeline
-# asr_pipe = pipeline("automatic-speech-recognition", model="Abdullah17/whisper-small-urdu")
-# from difflib import SequenceMatcher
+# Capture audio samples from the microphone
+def capture_audio(rec_duration=6, sample_rate=16000):
+    audio_data = sd.rec(int(rec_duration * sample_rate), samplerate=sample_rate, channels=1)
+    sd.wait()
+    return audio_data.flatten()
 
-# # List of commands
-# commands = [
-#     "نمائندے ایجنٹ نمائندہ",
-#     "  سم  ایکٹیویٹ ",
-#     " سم  بلاک بند ",
-#     "موبائل پیکیجز انٹرنیٹ پیکیج",
-#     " چالان جمع ",
-#     " گانا سنانا"
-# ]
-# # replies = [
-# # 1,2,
-# # ]
-# # Function to find the most similar command
 def find_most_similar_command(statement, command_list):
     best_match = None
     highest_similarity = 0
@@ -62,14 +33,10 @@ def find_most_similar_command(statement, command_list):
         i+=1
 
     return best_match,reply
-# x
-# get_text_from_voice("urdu.wav")
-import gradio as gr
-
 
 iface = gr.Interface(
     fn=transcribe_the_command,
-    inputs=gr.inputs.Audio(label="Recorded Audio",source="microphone"),
+    inputs=gr.inputs.Function(capture_audio, label="Recorded Audio"),
     outputs="text",
     title="Whisper Small Urdu Command",
     description="Realtime demo for Urdu speech recognition using a fine-tuned Whisper small model and outputting the estimated command on the basis of speech transcript.",