Update app.py

app.py

@@ -1,34 +1,31 @@
 import gradio as gr
-from transformers import BartTokenizer, BartForConditionalGeneration
-import whisper
+import requests
 
-# Initialize the BART model and tokenizer
-MODEL_NAME = "facebook/bart-large-cnn"
-model = BartForConditionalGeneration.from_pretrained(MODEL_NAME)
-tokenizer = BartTokenizer.from_pretrained(MODEL_NAME)
+API_URL = "https://api-inference.huggingface.co/models/openai/whisper-large-v2/whisper"
+API_KEY = "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"
+HEADERS = {"Authorization": f"Bearer {API_KEY}"}
 
-def convert_and_summarize(audio_path: str) -> str:
-    # Convert audio to text
-    whisper_model = whisper.load_model("base")
-    result = whisper_model.transcribe(audio_path)
-    transcribed_text = result["text"]
+def transcribe_audio(audio_path: str) -> str:
+    # Read audio file
+    with open(audio_path, "rb") as f:
+        audio_data = f.read()
     
-    # Summarize the transcribed text
-    inputs = tokenizer([transcribed_text], max_length=1024, truncation=True, return_tensors='pt')
-    summary_ids = model.generate(inputs['input_ids'])
-    summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
+    # Make API request to OpenAI Whisper v2 API
+    response = requests.post(API_URL, headers=HEADERS, data=audio_data)
+    result = response.json()
+    transcribed_text = result["text"]
     
-    return summary
+    return transcribed_text
 
 audio_input = gr.inputs.Audio(type="filepath")
+text_output = gr.outputs.Textbox()
 
-# Interface for Gradio
 iface = gr.Interface(
-    fn=convert_and_summarize,
+    fn=transcribe_audio,
     inputs=audio_input,
-    outputs="text",
-    title="Audio-to-Summarized-Text",
-    description="Upload an audio here and get a bullet-point summary of its content.",
+    outputs=text_output,
+    title="Speech-to-Text using Whisper v2",
+    description="Upload an audio file to transcribe it to text.",
     theme="Monochrome",
     live=True,
     capture_session=True,