vibecoding/podx
1
0
Fork 0
Code Issues 4 Pull Requests Actions Packages Projects Releases Wiki Activity

Fix whisper auto bug

Browse Source
This commit is contained in:
Tomas Kracmar
2025-09-07 13:45:00 +02:00
parent efe166a4ff
commit 420442506a
2 changed files with 12 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
app/worker.py
View File

@@ -9,6 +9,7 @@ TRN = Path(os.getenv("TRANSCRIPT_ROOT", "/transcripts"))
TMP = Path(os.getenv("TMP_ROOT", "/tmpdl")) TMP = Path(os.getenv("TMP_ROOT", "/tmpdl"))
MODEL_NAME = os.getenv("WHISPER_MODEL","large-v3") MODEL_NAME = os.getenv("WHISPER_MODEL","large-v3")
COMPUTE = os.getenv("WHISPER_PRECISION","int8") COMPUTE = os.getenv("WHISPER_PRECISION","int8")
WHISPER_LANGUAGE = os.getenv("WHISPER_LANGUAGE", "auto").strip()
OWUI_URL = os.getenv("OPENWEBUI_URL", "").rstrip("/") OWUI_URL = os.getenv("OPENWEBUI_URL", "").rstrip("/")
OWUI_KEY = os.getenv("OPENWEBUI_API_KEY", "") OWUI_KEY = os.getenv("OPENWEBUI_API_KEY", "")
@@ -78,7 +79,8 @@ def yt_dlp(url, outdir):
def transcribe(media_path: Path): def transcribe(media_path: Path):
model = get_model() model = get_model()
segments, info = model.transcribe(str(media_path), vad_filter=True, language="auto") lang = None if WHISPER_LANGUAGE.lower() == "auto" else WHISPER_LANGUAGE
segments, info = model.transcribe(str(media_path), vad_filter=True, language=lang)
title = media_path.stem title = media_path.stem
base = TRN / title base = TRN / title
segs, text_parts = [], [] segs, text_parts = [], []
@@ -231,8 +233,6 @@ def publish_to_openwebui(paths):
except Exception as e: except Exception as e:
log({"status": "owui_error", "error": str(e)}) log({"status": "owui_error", "error": str(e)})
from pathlib import Path
def handle_local_file(path_str: str): def handle_local_file(path_str: str):
"""Transcribe & index a local media file that already exists in /library. """Transcribe & index a local media file that already exists in /library.
Safe to call repeatedly; it skips if transcript JSON already exists. Safe to call repeatedly; it skips if transcript JSON already exists.

9
docker-compose.yml
View File

@@ -21,6 +21,11 @@ services:
restart: unless-stopped restart: unless-stopped
extra_hosts: extra_hosts:
- host.docker.internal:host-gateway - host.docker.internal:host-gateway
healthcheck:
test: ["CMD-SHELL", "curl -fsS http://127.0.0.1:8080/health || exit 1"]
interval: 30s
timeout: 5s
retries: 3
podx-worker: podx-worker:
build: ./app build: ./app
@@ -43,6 +48,8 @@ services:
- ${MODELS_HOST_DIR:-./models}:/root/.cache/huggingface - ${MODELS_HOST_DIR:-./models}:/root/.cache/huggingface
depends_on: [meili, redis] depends_on: [meili, redis]
restart: unless-stopped restart: unless-stopped
healthcheck:
test: ["CMD-SHELL", "exit 0"]
extra_hosts: extra_hosts:
- host.docker.internal:host-gateway - host.docker.internal:host-gateway
@@ -100,4 +107,6 @@ services:
- ${LIBRARY_HOST_DIR:-./library}:/library - ${LIBRARY_HOST_DIR:-./library}:/library
- ${TRANSCRIPTS_HOST_DIR:-./transcripts}:/transcripts - ${TRANSCRIPTS_HOST_DIR:-./transcripts}:/transcripts
depends_on: [redis] depends_on: [redis]
healthcheck:
test: ["CMD-SHELL", "exit 0"]
restart: unless-stopped restart: unless-stopped