Update app.py

app.py

@@ -1,20 +1,20 @@
-# Install necessary libraries (if not installed)
-# !pip install gradio transformers soundfile torch
+# Install necessary libraries (if not already installed)
+!pip install gradio transformers soundfile torch
 
-import gradio as gr
 import torch
 import soundfile as sf
+import gradio as gr
 from transformers import SpeechT5ForTextToSpeech, SpeechT5Processor, SpeechT5HifiGan
 
-# Load the pre-trained model, vocoder, and processor
+# Load your fine-tuned model, processor, and vocoder
 model = SpeechT5ForTextToSpeech.from_pretrained("krishna195/speecht5_krishna_finatuned")
 vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan")
 processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
 
-# Dummy speaker embeddings for speech generation (replace with actual embeddings if needed)
-speaker_embeddings = torch.randn(1, 512)  # Example speaker embedding size
+# Use pre-defined speaker embeddings (you can replace this with your actual embeddings)
+speaker_embeddings = torch.randn(1, 512)  # Example embedding size, adjust to your speaker embeddings
 
-# Function to generate speech from input text
+# Function to generate speech from text
 def text_to_speech(input_text):
     # Process the input text
     inputs = processor(text=input_text, return_tensors="pt")
@@ -22,23 +22,26 @@ def text_to_speech(input_text):
     # Generate speech using the model and vocoder
     speech = model.generate_speech(inputs["input_ids"], speaker_embeddings, vocoder=vocoder)
 
-    # Save the audio to a file (temporary storage)
+    # Save the generated speech to a temporary file
     output_file = "generated_speech.wav"
     sf.write(output_file, speech.numpy(), 16000)
 
-    # Return the path to the audio file for Gradio to play it
+    # Return the path to the audio file for Gradio to play
     return output_file
 
-# Create Gradio UI
+# Create the Gradio UI interface
 iface = gr.Interface(
     fn=text_to_speech,
-    inputs=gr.Textbox(label="Enter Text"),
+    inputs="text",
     outputs="audio",
-    title="Text to Speech Generator",
-    description="Enter the text you want to convert to speech, and the model will generate the corresponding audio.",
-    layout="vertical",  # Use vertical layout
-    theme="default"  # Use default theme
+    title="Text to Speech Converter",
+    description="Enter text and convert it into speech using a fine-tuned SpeechT5 model.",
+    examples=[
+        ["Hello, how are you doing today?"],
+        ["Speech synthesis is amazing with deep learning models."],
+        ["TensorFlow and PyTorch are powerful machine learning frameworks."]
+    ]
 )
 
-# Launch the Gradio interface
+# Launch the Gradio app
 iface.launch()