Create app.py

app.py

@@ -0,0 +1,56 @@
+import gradio as gr
+import whisper
+import tempfile
+import os
+
+model = whisper.load_model("small")  # supports 50+ languages
+
+def transcribe(audio):
+    if audio is None:
+        return {"error": "no audio"}
+
+    # audio = (sample_rate, numpy_array)
+    sample_rate, data = audio
+
+    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp:
+        import soundfile as sf
+        sf.write(tmp.name, data, sample_rate)
+        path = tmp.name
+
+    result = model.transcribe(
+        path,
+        word_timestamps=True,
+        verbose=False
+    )
+
+    os.remove(path)
+
+    segments = []
+    for seg in result["segments"]:
+        segments.append({
+            "start": round(seg["start"], 2),
+            "end": round(seg["end"], 2),
+            "text": seg["text"].strip(),
+            "words": [
+                {
+                    "word": w["word"],
+                    "start": round(w["start"], 2),
+                    "end": round(w["end"], 2)
+                }
+                for w in seg.get("words", [])
+            ]
+        })
+
+    return {
+        "language": result["language"],
+        "segments": segments
+    }
+
+iface = gr.Interface(
+    fn=transcribe,
+    inputs=gr.Audio(type="numpy"),
+    outputs="json",
+    allow_flagging="never"
+)
+
+iface.launch(server_name="0.0.0.0")