diff --git a/examples/online_serving/text_to_speech/qwen3_tts/gradio_demo.py b/examples/online_serving/text_to_speech/qwen3_tts/gradio_demo.py
index 8b39b9eab..3c99a82d3 100644
--- a/examples/online_serving/text_to_speech/qwen3_tts/gradio_demo.py
+++ b/examples/online_serving/text_to_speech/qwen3_tts/gradio_demo.py
@@ -408,7 +408,10 @@ def create_app(api_base: str):
                 await resp.aclose()
                 await client.aclose()
 
-        return StreamingResponse(relay(), media_type="application/octet-stream")
+        # Use audio/pcm to bypass browsers (e.g. Chromium) MIME type sniffing, which
+        # buffers application/octet-stream responses before delivering chunks to
+        # fetch().body.getReader(), inflating TTFP to total_ms.
+        return StreamingResponse(relay(), media_type="audio/pcm")
 
     # ── Gradio UI ────────────────────────────────────────────────
     voices = fetch_voices(api_base)