Update app.py

app.py

@@ -1,72 +1,72 @@
-import gradio as gr
-from transformers import pipeline
-import numpy as np
-import librosa
-import pandas as pd
-
-
-MODEL_NAME = "openai/whisper-tiny"
-BATCH_SIZE = 8
-# device = 0 if torch.cuda.is_available() else "cpu"
-
-pipe = pipeline(
-    task="automatic-speech-recognition",
-    model=MODEL_NAME,
-    chunk_length_s=30,
-    # device=device,
-)
-
-# eng_classifier = pipeline("text-classification", model="Hate-speech-CNERG/bert-base-uncased-hatexplain")
-
-def format_output_to_list(data):
-    formatted_list = "\n".join([f"{item['timestamp'][0]}s - {item['timestamp'][1]}s \t : {item['text']}" for item in data])
-    return formatted_list
-
-def transcribe(inputs, task):
-    if inputs is None:
-        raise gr.Error("No audio file submitted! Please upload or record an audio file before submitting your request.")
-
-    output = pipe(inputs, batch_size=BATCH_SIZE, return_timestamps="word", generate_kwargs={"task": task})
-    text = output['text']
-    timestamps = format_output_to_list(output['chunks'])
-    return  [text, timestamps]
-
-examples = [
-        ["arabic_english_audios/audios/arabic_audio_1.wav"],
-        ["arabic_english_audios/audios/arabic_audio_2.wav"],
-        ["arabic_english_audios/audios/arabic_audio_3.wav"],
-        ["arabic_english_audios/audios/arabic_audio_4.wav"],
-        ["arabic_english_audios/audios/arabic_hate_audio_1.mp3"],
-        ["arabic_english_audios/audios/arabic_hate_audio_2.mp3"],
-        ["arabic_english_audios/audios/arabic_hate_audio_3.mp3"],
-        ["arabic_english_audios/audios/english_audio_1.wav"],
-        ["arabic_english_audios/audios/english_audio_2.mp3"],
-        ["arabic_english_audios/audios/english_audio_3.mp3"],
-        ["arabic_english_audios/audios/english_audio_4.mp3"],
-        ["arabic_english_audios/audios/english_audio_5.mp3"],
-        ["arabic_english_audios/audios/english_audio_6.wav"]
-    ]
-
-with gr.Blocks(theme=gr.themes.Default()) as demo:
-    gr.HTML("<h1 style='text-align: center;'>Transcribe Audio with Timestamps using whisper-large-v3</h1>")
-    gr.Markdown("")
-    with gr.Row():
-        with gr.Column():
-            audio_input = gr.Audio(sources=["upload", 'microphone'], type="filepath", label="Audio file")
-            task = gr.Radio(["transcribe", "translate"], label="Task")
-            with gr.Row():
-                clear_button = gr.ClearButton(value="Clear")
-                submit_button = gr.Button("Submit", variant="primary", )
-            
-        with gr.Column():
-            transcript_output = gr.Text(label="Transcript")
-            timestamp_output = gr.Text(label="Timestamp")
-    
-    examples = gr.Examples(examples, inputs=audio_input, outputs=[transcript_output, timestamp_output], fn=transcribe, examples_per_page=20)
-
-    submit_button.click(fn=transcribe, inputs=audio_input, outputs=[transcript_output, timestamp_output])
-    clear_button.add([audio_input, transcript_output, timestamp_output])
-
-
-if __name__ == "__main__":
-    demo.launch()
+import gradio as gr
+from transformers import pipeline
+import numpy as np
+import librosa
+import pandas as pd
+
+
+MODEL_NAME = "openai/whisper-large-v3"
+BATCH_SIZE = 8
+# device = 0 if torch.cuda.is_available() else "cpu"
+
+pipe = pipeline(
+    task="automatic-speech-recognition",
+    model=MODEL_NAME,
+    chunk_length_s=30,
+    # device=device,
+)
+
+# eng_classifier = pipeline("text-classification", model="Hate-speech-CNERG/bert-base-uncased-hatexplain")
+
+def format_output_to_list(data):
+    formatted_list = "\n".join([f"{item['timestamp'][0]}s - {item['timestamp'][1]}s \t : {item['text']}" for item in data])
+    return formatted_list
+
+def transcribe(inputs, task):
+    if inputs is None:
+        raise gr.Error("No audio file submitted! Please upload or record an audio file before submitting your request.")
+
+    output = pipe(inputs, batch_size=BATCH_SIZE, return_timestamps="word", generate_kwargs={"task": task})
+    text = output['text']
+    timestamps = format_output_to_list(output['chunks'])
+    return  [text, timestamps]
+
+examples = [
+        ["arabic_english_audios/audios/arabic_audio_1.wav"],
+        ["arabic_english_audios/audios/arabic_audio_2.wav"],
+        ["arabic_english_audios/audios/arabic_audio_3.wav"],
+        ["arabic_english_audios/audios/arabic_audio_4.wav"],
+        ["arabic_english_audios/audios/arabic_hate_audio_1.mp3"],
+        ["arabic_english_audios/audios/arabic_hate_audio_2.mp3"],
+        ["arabic_english_audios/audios/arabic_hate_audio_3.mp3"],
+        ["arabic_english_audios/audios/english_audio_1.wav"],
+        ["arabic_english_audios/audios/english_audio_2.mp3"],
+        ["arabic_english_audios/audios/english_audio_3.mp3"],
+        ["arabic_english_audios/audios/english_audio_4.mp3"],
+        ["arabic_english_audios/audios/english_audio_5.mp3"],
+        ["arabic_english_audios/audios/english_audio_6.wav"]
+    ]
+
+with gr.Blocks(theme=gr.themes.Default()) as demo:
+    gr.HTML("<h1 style='text-align: center;'>Transcribe Audio with Timestamps using whisper-large-v3</h1>")
+    gr.Markdown("")
+    with gr.Row():
+        with gr.Column():
+            audio_input = gr.Audio(sources=["upload", 'microphone'], type="filepath", label="Audio file")
+            task = gr.Radio(["transcribe", "translate"], label="Task")
+            with gr.Row():
+                clear_button = gr.ClearButton(value="Clear")
+                submit_button = gr.Button("Submit", variant="primary", )
+            
+        with gr.Column():
+            transcript_output = gr.Text(label="Transcript")
+            timestamp_output = gr.Text(label="Timestamp")
+    
+    examples = gr.Examples(examples, inputs=audio_input, outputs=[transcript_output, timestamp_output], fn=transcribe, examples_per_page=20)
+
+    submit_button.click(fn=transcribe, inputs=audio_input, outputs=[transcript_output, timestamp_output])
+    clear_button.add([audio_input, transcript_output, timestamp_output])
+
+
+if __name__ == "__main__":
+    demo.launch()