Create app.py

app.py

@@ -0,0 +1,31 @@
+import gradio as gr
+from transformers import VitsModel, AutoTokenizer
+import torch
+import scipy.io.wavfile
+import tempfile
+
+# Load the Somali TTS model
+model = VitsModel.from_pretrained("facebook/mms-tts-som")
+tokenizer = AutoTokenizer.from_pretrained("facebook/mms-tts-som")
+
+def somali_text_to_speech(text):
+    inputs = tokenizer(text, return_tensors="pt")
+
+    with torch.no_grad():
+        output = model(**inputs)
+    
+    waveform = output.waveform.squeeze().cpu().numpy()
+
+    # Save waveform to a temporary WAV file
+    with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as tmp:
+        scipy.io.wavfile.write(tmp.name, rate=model.config.sampling_rate, data=waveform)
+        return tmp.name
+
+# Launch Gradio Interface
+gr.Interface(
+    fn=somali_text_to_speech,
+    inputs=gr.Textbox(label="Enter Somali Text"),
+    outputs=gr.Audio(label="Generated Somali Speech"),
+    title="Somali Text-to-Speech",
+    description="Type Somali text and hear it spoken using Hugging Face's VitsModel."
+).launch(share=True)