Update app.py

app.py

@@ -1,5 +1,5 @@
 import gradio as gr
-from transformers import pipeline
+from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
 import numpy as np
 import pandas as pd
 import re
@@ -8,17 +8,29 @@ from pydub.generators import Sine
 import io
 import torch
 
-MODEL_NAME = "openai/whisper-large-v3"
-BATCH_SIZE = 8
-device = 0 if torch.cuda.is_available() else "cpu"
+device = "cuda:0" if torch.cuda.is_available() else "cpu"
+model_id = "openai/whisper-large-v3"
+
+model = AutoModelForSpeechSeq2Seq.from_pretrained(
+    model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True
+)
+model.to(device)
+
+processor = AutoProcessor.from_pretrained(model_id)
 
 pipe = pipeline(
-    task="automatic-speech-recognition",
-    model=MODEL_NAME,
+    "automatic-speech-recognition",
+    model=model,
+    tokenizer=processor.tokenizer,
+    feature_extractor=processor.feature_extractor,
+    max_new_tokens=128,
     chunk_length_s=30,
+    batch_size=8,
+    torch_dtype=torch_dtype,
     device=device,
 )
 
+
 arabic_bad_Words = pd.read_csv("arabic_bad_words_dataset.csv")
 english_bad_Words = pd.read_csv("english_bad_words_dataset.csv")
 
@@ -127,10 +139,10 @@ examples = [
         ["arabic_english_audios/audios/english_audio_34.mp3", 'English', 'transcribe', 'word'],
         ["arabic_english_audios/audios/english_audio_35.mp3", 'English', 'transcribe', 'word'],
         ["arabic_english_audios/audios/english_audio_36.mp3", 'English', 'transcribe', 'word'],
-        ["arabic_english_audios/audios/english_audio_37.mp3", 'English', 'transcribe', 'word'],
-        ["arabic_english_audios/audios/english_audio_38.mp3", 'English', 'transcribe', 'word'],
-        ["arabic_english_audios/audios/english_audio_39.mp3", 'English', 'transcribe', 'word'],
-        ["arabic_english_audios/audios/english_audio_40.mp3", 'English', 'transcribe', 'word'],
+        # ["arabic_english_audios/audios/english_audio_37.mp3", 'English', 'transcribe', 'word'],
+        # ["arabic_english_audios/audios/english_audio_38.mp3", 'English', 'transcribe', 'word'],
+        # ["arabic_english_audios/audios/english_audio_39.mp3", 'English', 'transcribe', 'word'],
+        # ["arabic_english_audios/audios/english_audio_40.mp3", 'English', 'transcribe', 'word'],
         # ["arabic_english_audios/audios/english_audio_41.mp3", 'English', 'transcribe', 'word'],
         # ["arabic_english_audios/audios/english_audio_42.mp3", 'English', 'transcribe', 'word'],
         # ["arabic_english_audios/audios/english_audio_43.mp3", 'English', 'transcribe', 'word'],