add sentences borders enable in pc player

sentence_api/main.py

@@ -2,6 +2,7 @@ import hashlib
 import hmac
 import logging
 import mimetypes
+import tempfile
 import uuid
 import urllib.request
 import urllib.error
@@ -57,7 +58,7 @@ from .models import (
     VideoUploadResponse,
 )
 from .oss import VolcanoOSSUploader
-from .processing import VideoProcessor, split_sentence_text
+from .processing import VideoProcessor, generate_audio_boundary_document, split_sentence_text
 from .repository import VideoRepository
 from .store import BoundaryStore
 from .transcription import MossTranscriber, Transcriber
@@ -72,6 +73,7 @@ SHA256_PATH = ApiPath(
 )
 ALLOWED_VIDEO_EXTENSIONS = {".mp4", ".m4v", ".mov", ".mkv", ".webm"}
 ALLOWED_AUDIO_EXTENSIONS = {".m4a", ".mp4", ".wav"}
+MAX_AUDIO_BOUNDARY_BYTES = 512 * 1024 * 1024
 
 
 class UploadTooLargeError(ValueError):
@@ -329,6 +331,47 @@ def create_app(
             raise HTTPException(status_code=503, detail=str(exc)) from exc
         return SentenceExplanationResponse(explanation=explanation)
 
+    @application.post(
+        "/api/v1/audio/sentence-boundaries",
+        response_model=SentenceBoundaryDocument,
+    )
+    async def generate_sentence_boundaries(
+        file: UploadFile = File(...),
+        language: Optional[str] = Form(default=None, max_length=32),
+        user: AuthenticatedUser = Depends(current_user),
+    ) -> SentenceBoundaryDocument:
+        del user
+        max_bytes = min(MAX_AUDIO_BOUNDARY_BYTES, service_settings.max_upload_bytes)
+        with tempfile.TemporaryDirectory(prefix="audio-boundaries-") as temp_dir:
+            audio_path = Path(temp_dir) / "audio.wav"
+            try:
+                audio_hash, _size = await run_in_threadpool(
+                    _persist_upload,
+                    file.file,
+                    audio_path,
+                    max_bytes,
+                )
+                document = await run_in_threadpool(
+                    generate_audio_boundary_document,
+                    video_hash=audio_hash,
+                    audio_path=audio_path,
+                    transcriber=moss,
+                    language=language,
+                )
+            except UploadTooLargeError as exc:
+                raise HTTPException(status_code=413, detail=str(exc)) from exc
+            except ValueError as exc:
+                raise HTTPException(status_code=422, detail=str(exc)) from exc
+            except Exception as exc:
+                logger.exception("Ad-hoc audio sentence boundary generation failed.")
+                raise HTTPException(
+                    status_code=502,
+                    detail="Sentence boundary generation failed. Please try again later.",
+                ) from exc
+            finally:
+                await file.close()
+            return document
+
     @application.delete("/api/v1/courses/{video_hash}/enroll", status_code=204)
     def unenroll_course(
         video_hash: str = SHA256_PATH,