Merge pull request #10 from kanjieater/video

adding video syncing

.dockerignore

@@ -1,6 +1,7 @@
 .env
 __pycache__/
 build/
+.ruff_cache/
 .vscode/
 *.egg-info
 

.gitignore

@@ -1,5 +1,6 @@
 .env
 __pycache__/
+.ruff_cache/
 .vscode/
 build/
 *.egg-info

.vscode/launch.json

@@ -16,6 +16,22 @@
         // "/mnt/d/sync/変な家/変な家.epub",
       ],
     },
+    {
+      "name": "Video file",
+      "type": "debugpy",
+      "request": "launch",
+      "module": "subplz",
+      "args": [
+        "sync",
+        "-d",
+        "/mnt/v/test",
+        "--no-overwrite",
+        // "--audio",
+        // "/mnt/d/sync/変な家/変な家.m4b",
+        // "--text",
+        // "/mnt/d/sync/変な家/変な家.epub",
+      ],
+    },
 
     {
       "name": "Python: align vtt",

Dockerfile

@@ -17,9 +17,6 @@ RUN apt-get update \
         libportaudio2 \
     && rm -rf /var/lib/apt/lists/*
 
-#remove when we get abts updated w/ a pyproject.toml
-COPY /home/ke/code/AudiobookTextSync /tmp/AudiobookTextSync
-
 COPY pyproject.toml /tmp/
 RUN pip install --no-cache-dir /tmp/
 
@@ -36,5 +33,5 @@ WORKDIR /app
 HEALTHCHECK --interval=30s --timeout=10s CMD nc -z localhost 10300 || exit 1
 
 # Start the application
-# ENTRYPOINT ["python", "./src/main.py"]
-# CMD ["--help"]
+ENTRYPOINT ["python", "-m", "subplz"]
+CMD ["--help"]

compose.yml

@@ -10,7 +10,7 @@ services:
       - TZ=Etc/UTC
       - WHISPER_MODEL=tiny-int8
       - WHISPER_BEAM=1 #optional
-      - WHISPER_LANG=en #optional
+      - WHISPER_LANG=ja #optional
     volumes:
       - ./:/app
       - /mnt/d/sync:/sync

readme.md

@@ -59,6 +59,14 @@ Currently supports Docker (preferred), Windows, and unix based OS's like Ubuntu
 - This can be GPU intense, RAM intense, and CPU intense script part. `subplz sync -d "<full folder path>"` eg `subplz sync -d "/mnt/d/Editing/Audiobooks/かがみの孤城/"`. This runs each file to get a character level transcript. It then creates a sub format that can be matched to the `script.txt`. Each character level subtitle is merged into a phrase level, and your result should be a `<name>.srt` file. The video or audio file then can be watched with `MPV`, playing audio in time with the subtitle.
 - Users with a modern CPU with lots of threads won't notice much of a difference between using CUDA/GPU & CPU
 
+# Overwrite
+By default the tool will overwrite any existing srt named after the audio file's name. If you don't want it to do this you must explicitly tell it not to.
+
+`subplz sync -d "/mnt/v/somefolder" --no-overwrite`
+
+# Only Running for the Files It Needs
+SubPlz writes a file in the same folder to the audio with the `<audiofile>.subplz` extension. It starts with a `.` so you may need to enable hidden files if you want to see it. This ensures that subplz runs once and only once per directory for your content. If you want to rerun the SubPlz syncing, delete the file.
+
 
 # Split m4b by chapter
 `./split.sh "/mnt/d/Editing/Audiobooks/かがみの孤城/"`

subplz/cli.py

@@ -69,7 +69,7 @@ def setup_advanced_cli(parser):
     # Behaviors
     optional_group.add_argument(
         "--overwrite",
-        default=False,
+        default=True,
         help="Overwrite any destination files",
         action=argparse.BooleanOptionalAction,
     )