fix: add transcription model fallback chain

openrouter_client.py

@@ -36,17 +36,22 @@ def _audio_format(audio_path: str) -> str:
     return suffix or "wav"
 
 
-def _build_transcription_payload(audio_path: str) -> dict[str, Any]:
+TRANSCRIPTION_MODELS = [
+    "openai/gpt-4o-mini-transcribe",
+    "openai/whisper-large-v3",
+]
+
+
+def _build_transcription_payload(audio_path: str, model: str) -> dict[str, Any]:
     with open(audio_path, "rb") as audio_file:
         encoded = base64.b64encode(audio_file.read()).decode("ascii")
 
     return {
-        "model": "openai/whisper-large-v3",
+        "model": model,
         "input_audio": {
             "data": encoded,
             "format": _audio_format(audio_path),
         },
-        "language": "en",
     }
 
 
@@ -84,31 +89,39 @@ async def _normalize_audio_for_transcription(audio_path: str) -> str:
 
 
 async def transcribe(audio_path: str) -> str:
-    """Send audio to OpenRouter's whisper model and return transcript text."""
+    """Send audio to OpenRouter STT models and return transcript text."""
     headers = _auth_headers()
     headers["Content-Type"] = "application/json"
+    headers["X-OpenRouter-Title"] = "discord-meeting-bot"
 
     normalized_path = await _normalize_audio_for_transcription(audio_path)
+    failures: list[str] = []
     try:
         async with httpx.AsyncClient(timeout=300) as client:
-            resp = await client.post(
-                f"{OPENROUTER_BASE}/audio/transcriptions",
-                headers=headers,
-                content=json.dumps(_build_transcription_payload(normalized_path)),
-            )
-            try:
-                resp.raise_for_status()
-            except httpx.HTTPStatusError as exc:
-                detail = summarize_error(_safe_json(resp), fallback=resp.text)
-                raise RuntimeError(
-                    f"OpenRouter transcription failed ({resp.status_code}): {detail}"
-                ) from exc
+            for model in TRANSCRIPTION_MODELS:
+                resp = await client.post(
+                    f"{OPENROUTER_BASE}/audio/transcriptions",
+                    headers=headers,
+                    json=_build_transcription_payload(normalized_path, model),
+                )
+                try:
+                    resp.raise_for_status()
+                except httpx.HTTPStatusError:
+                    detail = summarize_error(_safe_json(resp), fallback=resp.text)
+                    generation_id = resp.headers.get("x-generation-id")
+                    suffix = f"; generation_id={generation_id}" if generation_id else ""
+                    failures.append(f"{model}: HTTP {resp.status_code}: {detail}{suffix}")
+                    continue
 
-            data = resp.json()
-            text = data.get("text", "")
-            if not text.strip():
-                raise RuntimeError("OpenRouter transcription returned empty text")
-            return text.strip()
+                data = resp.json()
+                text = data.get("text", "")
+                if text.strip():
+                    return text.strip()
+                failures.append(f"{model}: returned empty text")
+
+        raise RuntimeError(
+            "OpenRouter transcription failed across all models: " + " | ".join(failures)
+        )
     finally:
         try:
             os.remove(normalized_path)

tests/test_openrouter_client.py

@@ -5,7 +5,7 @@ from pathlib import Path
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(__file__)))
 
-from openrouter_client import _audio_format, _build_transcription_payload
+from openrouter_client import TRANSCRIPTION_MODELS, _audio_format, _build_transcription_payload
 
 
 def test_audio_format_defaults_to_wav_when_missing_suffix(tmp_path: Path):
@@ -18,9 +18,8 @@ def test_build_transcription_payload_uses_base64_json_shape(tmp_path: Path):
     path = tmp_path / "meeting.wav"
     path.write_bytes(b"RIFFdemo")
 
-    payload = _build_transcription_payload(str(path))
+    payload = _build_transcription_payload(str(path), TRANSCRIPTION_MODELS[0])
 
-    assert payload["model"] == "openai/whisper-large-v3"
-    assert payload["language"] == "en"
+    assert payload["model"] == "openai/gpt-4o-mini-transcribe"
     assert payload["input_audio"]["format"] == "wav"
     assert payload["input_audio"]["data"] == base64.b64encode(b"RIFFdemo").decode("ascii")