feat: add whisper tasks

.env.test

@@ -1,4 +1,4 @@
 DATABASE_URI="postgresql://postgres:postgres@localhost:5432/whisperbox_test"
 ENVIRONMENT="development"
 API_SECRET="foo"
-BROKER_URI="redis://localhost:6379/0"
+BROKER_URL="redis://localhost:6379/0"

.flake8

@@ -1,4 +1,4 @@
 [flake8]
-max-line-length = 90
+max-line-length = 88
 extend-ignore = E203
 exclude = .git,__pycache__,__init__.py,.mypy_cache,.pytest_cache,app/alembic/versions

app/shared/celery.py

@@ -0,0 +1,9 @@
+from celery import Celery
+
+from app.shared.config import settings
+
+
+def get_celery_binding() -> Celery:
+    celery = Celery("tasks")
+    celery.conf.broker_url = settings.BROKER_URL
+    return celery

app/shared/config.py

@@ -9,10 +9,12 @@ class Settings(BaseSettings):
     ENVIRONMENT: str
 
     # derived settings
-    BROKER_URI: str
+    BROKER_URL: str
 
 
 if "pytest" in sys.modules:
-    settings = Settings(_env_file=".env.test", _env_file_encoding="utf-8")  # type: ignore
+    settings = Settings(
+        _env_file=".env.test", _env_file_encoding="utf-8"
+    )  # type: ignore
 else:
     settings = Settings()

app/shared/db/alembic/versions/426b6bdc3360_add_job_tables.py

@@ -1,8 +1,8 @@
 """add_job_tables
 
-Revision ID: bb249ed79907
+Revision ID: 426b6bdc3360
 Revises:
-Create Date: 2023-01-17 14:30:30.920466
+Create Date: 2023-01-27 17:55:21.758828
 
 """
 import sqlalchemy as sa
@@ -10,7 +10,7 @@ from alembic import op
 from sqlalchemy.dialects import postgresql
 
 # revision identifiers, used by Alembic.
-revision = "bb249ed79907"
+revision = "426b6bdc3360"
 down_revision = None
 branch_labels = None
 depends_on = None
@@ -23,12 +23,21 @@ def upgrade() -> None:
         sa.Column("url", sa.String(length=2048), nullable=True),
         sa.Column(
             "status",
-            sa.Enum("create", "error", "processing", "success", name="jobstatus"),
+            sa.Enum("create", "processing", "error", "success", name="jobstatus"),
             nullable=False,
         ),
-        sa.Column("type", sa.Enum("transcript", name="jobtype"), nullable=False),
+        sa.Column("config", sa.JSON(none_as_null=True), nullable=True),
+        sa.Column("meta", sa.JSON(none_as_null=True), nullable=True),
         sa.Column(
-            "created_at", sa.DateTime(), server_default=sa.text("now()"), nullable=False
+            "type",
+            sa.Enum("transcript", "translation", "language_detection", name="jobtype"),
+            nullable=False,
+        ),
+        sa.Column(
+            "created_at",
+            sa.DateTime(),
+            server_default=sa.text("now()"),
+            nullable=False,
         ),
         sa.Column("updated_at", sa.DateTime(), nullable=True),
         sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),
@@ -40,10 +49,15 @@ def upgrade() -> None:
         sa.Column("job_id", postgresql.UUID(as_uuid=True), nullable=False),
         sa.Column("data", sa.JSON(none_as_null=True), nullable=True),
         sa.Column(
-            "type", sa.Enum("raw_transcript", name="artifacttype"), nullable=False
+            "type",
+            sa.Enum("raw_transcript", name="artifacttype"),
+            nullable=False,
         ),
         sa.Column(
-            "created_at", sa.DateTime(), server_default=sa.text("now()"), nullable=False
+            "created_at",
+            sa.DateTime(),
+            server_default=sa.text("now()"),
+            nullable=False,
         ),
         sa.Column("updated_at", sa.DateTime(), nullable=True),
         sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),

app/shared/db/alembic/versions/684a5e546314_add_job_meta_field.py

@@ -1,27 +0,0 @@
-"""add_job_meta_field
-
-Revision ID: 684a5e546314
-Revises: bb249ed79907
-Create Date: 2023-01-18 13:38:07.692830
-
-"""
-import sqlalchemy as sa
-from alembic import op
-
-# revision identifiers, used by Alembic.
-revision = "684a5e546314"
-down_revision = "bb249ed79907"
-branch_labels = None
-depends_on = None
-
-
-def upgrade() -> None:
-    # ### commands auto generated by Alembic - please adjust! ###
-    op.add_column("jobs", sa.Column("meta", sa.JSON(none_as_null=True), nullable=True))
-    # ### end Alembic commands ###
-
-
-def downgrade() -> None:
-    # ### commands auto generated by Alembic - please adjust! ###
-    op.drop_column("jobs", "meta")
-    # ### end Alembic commands ###

app/shared/db/dtos.py

@@ -1,6 +1,6 @@
 import enum
 from datetime import datetime
-from typing import Any, Optional
+from typing import Any, List, Optional
 from uuid import UUID
 
 from pydantic import AnyHttpUrl, BaseModel
@@ -21,6 +21,8 @@ class ArtifactType(str, enum.Enum):
 
 class JobType(str, enum.Enum):
     transcript = "transcript"
+    translation = "translation"
+    language_detection = "language_detection"
 
 
 class JobStatus(str, enum.Enum):
@@ -30,8 +32,12 @@ class JobStatus(str, enum.Enum):
     success = "success"
 
 
-class JobMeta(BaseModel):
+class JobConfig(BaseModel):
     language: Optional[str]
+
+
+class JobMeta(BaseModel):
+    error: Optional[str]
     task_id: Optional[UUID]
 
 
@@ -40,10 +46,11 @@ class Job(WithDbFields):
     type: JobType
     url: AnyHttpUrl
     meta: Optional[JobMeta]
+    config: Optional[JobConfig]
 
 
 class Artifact(WithDbFields):
     # TODO: narrow type
-    data: Optional[Any]
+    data: Optional[List[Any]]
     job_id: UUID
     type: ArtifactType

app/shared/db/models.py

@@ -35,6 +35,7 @@ class Job(Base, WithStandardFields):
 
     url = Column(String(length=2048))
     status = Column(Enum(JobStatus), nullable=False)
+    config = Column(JSON(none_as_null=True))
     meta = Column(JSON(none_as_null=True))
     type = Column(Enum(JobType), nullable=False)
 

app/tests/test_api.py

@@ -86,7 +86,7 @@ def test_get_artifact_pass(
     auth_headers: Dict[str, str], db_session: Session, mock_job: models.Job
 ) -> None:
     artifact = models.Artifact(
-        data={}, job_id=mock_job.id, type=dtos.ArtifactType.raw_transcript
+        data=[], job_id=mock_job.id, type=dtos.ArtifactType.raw_transcript
     )
 
     db_session.add(artifact)

app/web/main.py

@@ -7,11 +7,12 @@ from sqlalchemy.orm import Session
 
 import app.shared.db.dtos as dtos
 import app.shared.db.models as models
+from app.shared.celery import get_celery_binding
 from app.shared.db.base import get_session
 from app.web.security import authenticate_api_key
-from app.worker.main import transcribe
 
 app = FastAPI()
+celery = get_celery_binding()
 
 api_router = APIRouter(prefix="/api/v1", dependencies=[Depends(authenticate_api_key)])
 
@@ -21,19 +22,33 @@ def api_root() -> Dict:
     return {}
 
 
-class TranscriptPayload(BaseModel):
+class PostJobPayload(BaseModel):
     url: AnyHttpUrl
     type: dtos.JobType
+    language: Optional[str]
 
 
 @api_router.post("/jobs", response_model=dtos.Job, status_code=201)
 def create_job(
-    payload: TranscriptPayload, session: Session = Depends(get_session)
+    payload: PostJobPayload,
+    session: Session = Depends(get_session),
 ) -> models.Job:
-    job = models.Job(url=payload.url, status=dtos.JobStatus.create, type=payload.type)
-    session.add(job)
-    session.flush()
+    # create a job with status "create" and save it to the database.
+    job = models.Job(
+        url=payload.url,
+        status=dtos.JobStatus.create,
+        type=payload.type,
+        config={"language": payload.language} if payload.language else None,
+    )
 
+    session.add(job)
+    session.commit()
+
+    # queue an async transcription task.
+    # we use a signature here to allow full separation of
+    # worker processes and dependencies.
+    transcribe = celery.signature("app.worker.main.transcribe")
+    # TODO: catch delivery errors.
     transcribe.delay(job.id)
 
     return job

app/worker/main.py

@@ -1,37 +1,55 @@
-from time import sleep
+from asyncio.log import logger
 from uuid import UUID
 
-from celery import Celery
 from sqlalchemy.orm import Session
 
 import app.shared.db.dtos as dtos
 import app.shared.db.models as models
-from app.shared.config import settings
+from app.shared.celery import get_celery_binding
 from app.shared.db.base import SessionLocal
+from app.worker.strategies.local import LocalStrategy
 
-celery = Celery(__name__)
-
-celery.conf.broker_url = settings.BROKER_URI
+celery = get_celery_binding()
 
 
-def update_job_status(db: Session, job_id: UUID, status: dtos.JobStatus) -> None:
-    db.begin()
-    job = db.query(models.Job).filter(models.Job.id == job_id).one()
+def update_job_status(db: Session, job: models.Job, status: dtos.JobStatus) -> None:
     job.status = status
     db.commit()
 
 
 @celery.task()
-def transcribe(job_id: UUID) -> int:
+def transcribe(job_id: UUID) -> None:
     try:
         db: Session = SessionLocal()
-        update_job_status(db, job_id, dtos.JobStatus.processing)
-        sleep(60)
-        update_job_status(db, job_id, dtos.JobStatus.success)
+        job = db.query(models.Job).filter(models.Job.id == job_id).one()
+
+        update_job_status(db, job, dtos.JobStatus.processing)
+
+        # pick a transcription strategy.
+        # currently only `local` is supported.
+        job_record = dtos.Job.from_orm(job)
+        strategy = LocalStrategy(
+            db=db, job_id=job.id, url=job_record.url, config=job_record.config
+        )
+
+        # process selected task.
+        # currently only `transcribe` is supported.
+        if job.type == dtos.JobType.transcript:
+            result = strategy.transcribe()
+        elif job.type == dtos.JobType.translation:
+            result = strategy.translate()
+        else:
+            result = strategy.detect_language()
+
+        artifact = models.Artifact(
+            job_id=job.id, data=result, type=dtos.ArtifactType.raw_transcript
+        )
+        db.add(artifact)
         db.commit()
-    except Exception:
-        update_job_status(db, job_id, dtos.JobStatus.error)
+
+        update_job_status(db, job, dtos.JobStatus.success)
+    except Exception as e:
+        logger.error(e)
+        update_job_status(db, job, dtos.JobStatus.error)
     finally:
         db.close()
-
-    return 0

app/worker/strategies/local.py

@@ -0,0 +1,81 @@
+import os
+import shutil
+import tempfile
+from os import path
+from typing import Any, List, Literal, Optional
+from uuid import UUID
+
+import requests
+from pydantic import BaseModel
+from sqlalchemy.orm import Session
+from whisper import load_model
+
+import app.shared.db.dtos as dtos
+
+
+class DecodeOptions(BaseModel):
+    language: Optional[str]
+    task: Literal["translate", "transcribe"]
+
+
+class LocalStrategy:
+    def __init__(
+        self, db: Session, job_id: UUID, url: str, config: Optional[dtos.JobConfig]
+    ):
+        self.db = db
+        self.job_id = job_id
+        self.url = url
+        self.config = config
+
+    def transcribe(self) -> List[Any]:
+        result = self.run_whisper(self._download(), "transcribe")
+        self._cleanup()
+        return result
+
+    def translate(self) -> List[Any]:
+        result = self.run_whisper(self._download(), "translate")
+        self._cleanup()
+        return result
+
+    def detect_language(self) -> List[Any]:
+        raise NotImplementedError("detect_language has not been implemented yet.")
+
+    def _download(self) -> str:
+        dirname = self._get_tmp_dir()
+        filename = path.join(dirname, "media.mp3")
+
+        # re-create folder.
+        shutil.rmtree(dirname, ignore_errors=True)
+        os.makedirs(dirname)
+
+        # stream media to disk.
+        with requests.get(self.url, stream=True) as r:
+            r.raise_for_status()
+            with open(filename, "wb") as f:
+                for chunk in r.iter_content(chunk_size=8192):
+                    f.write(chunk)
+
+        return filename
+
+    def run_whisper(self, filepath: str, task: str) -> List[Any]:
+        language = self.config.language if self.config else None
+        decode_opts = DecodeOptions(task=task, language=language)
+        model = load_model("small", download_root="/models")
+
+        result = model.transcribe(
+            filepath, condition_on_previous_text=False, **decode_opts.dict()
+        )
+
+        return result["segments"]
+
+    def _get_tmp_dir(self) -> str:
+        return path.join(tempfile.gettempdir(), str(self.job_id))
+
+    def _cleanup(self) -> None:
+        shutil.rmtree(self._get_tmp_dir(), ignore_errors=True)
+
+    def _convert(self) -> None:
+        pass
+
+    def _transcribe(self) -> None:
+        pass

docker/app.dev.Dockerfile

@@ -1,17 +1,12 @@
 FROM python:3.10 AS compile-image
 
-COPY pyproject.toml .
-RUN --mount=type=cache,target=/root/.cache \
-    pip install --user .[web]
-
-FROM python:3.10 AS build-image
-
 WORKDIR /code
 
+COPY pyproject.toml .
+RUN pip install --no-cache-dir --user .[web]
+
 ENV PYTHONDONTWRITEBYTECODE 1
 ENV PYTHONUNBUFFERED 1
-
-COPY --from=compile-image /root/.local /root/.local
 ENV PATH=/root/.local/bin:$PATH
 
 ENTRYPOINT ["bash", "./app/web/start.sh"]

docker/dev.docker-compose.yml

@@ -4,7 +4,7 @@ x-app-variables: &app-variables
   API_SECRET: a_very_secret_token
   DATABASE_URI: postgresql://postgres:postgres@postgres/whisperbox
   ENVIRONMENT: development
-  BROKER_URI: redis://redis:6379/0
+  BROKER_URL: redis://redis:6379/0
 
 services:
   postgres:

docker/worker.dev.Dockerfile

@@ -1,20 +1,18 @@
 FROM python:3.10 AS compile-image
 
-RUN --mount=type=cache,target=/var/cache/apt \
-    apt-get update && apt-get install -y --no-install-recommends ffmpeg
-
-COPY pyproject.toml .
-RUN --mount=type=cache,target=/root/.cache \
-    pip install --user .[worker,worker_dev]
-
-FROM python:3.10 AS build-image
-
 WORKDIR /code
 
+RUN apt-get update && apt-get install -y --no-install-recommends ffmpeg
+
+COPY pyproject.toml .
+RUN pip install --no-cache-dir --user .[worker,worker_dev]
+
+COPY scripts/download_model.py .
+RUN chmod +x download_model.py && python download_model.py small small.en
+
+ENV PYTHONIOENCODING=utf-8
 ENV PYTHONDONTWRITEBYTECODE 1
 ENV PYTHONUNBUFFERED 1
-
-COPY --from=compile-image /root/.local /root/.local
 ENV PATH=/root/.local/bin:$PATH
 
-ENTRYPOINT ["watchmedo", "auto-restart", "-d" , "app/worker", "-p", "*.py", "celery", "--", "--app=app.worker.main.celery", "worker", "--loglevel=info"]
+ENTRYPOINT ["watchmedo", "auto-restart", "-d" , "app/worker", "-p", "*.py", "--recursive", "celery", "--", "--app=app.worker.main.celery", "worker", "--loglevel=info", "--concurrency=1"]

pyproject.toml

@@ -18,7 +18,8 @@ web=[
 ]
 
 worker=[
-  "whisper-openai ==1.0.0"
+  "whisper-openai ==1.0.0",
+  "requests ==2.28.2"
 ]
 
 lint = [

scripts/download_model.py

@@ -0,0 +1,7 @@
+import sys
+from whisper import _download, _MODELS # type: ignore
+
+if __name__ == "__main__":
+    args = sys.argv[1:]
+    for name in args:
+        _download(_MODELS[name], "/models/", False)