feat: add ASR provider support for voice co-creation

backend/app/services/adapters/asr/models.py

@@ -0,0 +1,11 @@
+"""ASR adapter result models."""
+
+from pydantic import BaseModel
+
+
+class TranscriptionOutput(BaseModel):
+    """Normalized speech-to-text output from one ASR provider."""
+
+    transcript_text: str
+    confidence: float | None = None
+    provider: str