Update app.py

app.py

@@ -139,12 +139,33 @@ def converti(path):
     summary = summarizer(abstr, max_length=56)
     summary_text = summary[0]['summary_text']
 
-    return summary_text
+    import torch
+    import soundfile as sf
+    from IPython.display import Audio
+    from datasets import load_dataset
+
+    from transformers import SpeechT5Processor, SpeechT5ForTextToSpeech
+
+    processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
+    model = SpeechT5ForTextToSpeech.from_pretrained("microsoft/speecht5_tts")
+
+    inputs = processor(text=summary_text, return_tensors="pt")
+
+    from datasets import load_dataset
+    embeddings_dataset = load_dataset("Matthijs/cmu-arctic-xvectors", split="validation")
+    speaker_embeddings = torch.tensor(embeddings_dataset[7306]["xvector"]).unsqueeze(0)
+
+    spectrogram = model.generate_speech(inputs["input_ids"], speaker_embeddings)
+
+    from transformers import SpeechT5HifiGan
+    vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan")
+    with torch.no_grad():
+        speech = vocoder(spectrogram)
+
+    speech = model.generate_speech(inputs["input_ids"], speaker_embeddings, vocoder=vocoder)
+    return Audio(speech, rate=16000)
     
 import gradio as gr
 
-#def greet(name):
-#    return "Hello " + name + "!!"
-#    return
-iface = gr.Interface(fn=converti, inputs=gr.Textbox(label="Input PDF name from your drive"), outputs="text")
+iface = gr.Interface(fn=converti, inputs=gr.Textbox(label="Input PDF name from your drive"), outputs="audio")
 iface.launch()