commit

app.py

@@ -0,0 +1,33 @@
+import gradio as gr
+import torch
+import scipy.io.wavfile as wavfile
+from transformers import pipeline
+
+# Load models
+asr = pipeline("automatic-speech-recognition", model="openai/whisper-small")
+tts = pipeline("text-to-speech", model="facebook/mms-tts-eng")
+
+def speech_to_speech(audio):
+    # Speech → Text
+    text = asr(audio)["text"]
+
+    # Text → Speech
+    speech = tts(text)
+
+    # Save audio
+    wavfile.write("output.wav", speech["sampling_rate"], speech["audio"])
+
+    return text, "output.wav"
+
+demo = gr.Interface(
+    fn=speech_to_speech,
+    inputs=gr.Audio(type="filepath", label="Speak here"),
+    outputs=[
+        gr.Textbox(label="Recognized Text"),
+        gr.Audio(label="Generated Speech")
+    ],
+    title="Speech to Speech AI",
+    description="Speak into the mic, AI listens and speaks back"
+)
+
+demo.launch()