omni-captions-skills 0.1.0__tar.gz → 0.1.2__tar.gz

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/.github/workflows/test.yml

@@ -33,14 +33,16 @@ jobs:
       - name: Install dependencies
         run: |
           pip install --upgrade pip
-          pip install -e ".[all]"
+          pip install -e ".[all]" --extra-index-url https://lattifai.github.io/pypi/simple
 
       - name: Test CLI entry point
+        env:
+          LATTIFAI_API_KEY: ${{ secrets.LATTIFAI_API_KEY }}
         run: |
           python -m omnicaptions --help
           omnicaptions --help
           # test LaiCut
-          lai alignment align tests/data/SA1.mp3 tests/data/SA1.vtt tests/data/SA1_LaiCut.vtt
+          lai alignment align -Y tests/data/SA1.mp3 tests/data/SA1.vtt tests/data/SA1_LaiCut.vtt
 
       - name: Run tests
         run: pytest tests/ -v

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/.gitignore

@@ -210,3 +210,4 @@ __marimo__/
 .worktrees/
 
 .DS_Store
+tests/data/SA1_LaiCut.vtt

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: omni-captions-skills
-Version: 0.1.0
+Version: 0.1.2
 Summary: AI-powered media transcription with Claude Code skills
 Author: LattifAI
 License-Expression: MIT
@@ -20,12 +20,14 @@ Requires-Dist: google-genai>=1.0.0
 Requires-Dist: lattifai-captions>=0.1.6
 Requires-Dist: yt-dlp>=2025.12.08
 Provides-Extra: all
-Requires-Dist: omnicaptions[dev,laicut]; extra == 'all'
+Requires-Dist: lattifai[diarization]>=1.3.4; extra == 'all'
+Requires-Dist: pytest>=8.0.0; extra == 'all'
+Requires-Dist: ruff>=0.4.0; extra == 'all'
 Provides-Extra: dev
 Requires-Dist: pytest>=8.0.0; extra == 'dev'
 Requires-Dist: ruff>=0.4.0; extra == 'dev'
 Provides-Extra: laicut
-Requires-Dist: lattifai[diarization]>=1.3.0; extra == 'laicut'
+Requires-Dist: lattifai[diarization]>=1.3.4; extra == 'laicut'
 Description-Content-Type: text/markdown
 
 # omni-captions-skills
@@ -94,6 +96,15 @@ Vibe coding 就是不用自己写代码的编程方式。
 
 Standard transcription gives "approximate" timestamps. LaiCut uses [LattifAI](https://lattifai.com/) Lattice-1 model to match text precisely to audio waveforms, achieving **word-level accuracy**.
 
+**Install LaiCut:**
+```bash
+# Using uv (recommended, auto-configures package index)
+uv pip install "omni-captions-skills[laicut]" --extra-index-url https://lattifai.github.io/pypi/simple/
+
+# Using pip
+pip install "omni-captions-skills[laicut]" --extra-index-url https://lattifai.github.io/pypi/simple/
+```
+
 **Supported languages:** English, Chinese, German, and mixed
 
 **Recommended workflow:** Align before translate (translated text doesn't match original audio)

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/README.md

@@ -64,6 +64,15 @@ Vibe coding 就是不用自己写代码的编程方式。
 
 Standard transcription gives "approximate" timestamps. LaiCut uses [LattifAI](https://lattifai.com/) Lattice-1 model to match text precisely to audio waveforms, achieving **word-level accuracy**.
 
+**Install LaiCut:**
+```bash
+# Using uv (recommended, auto-configures package index)
+uv pip install "omni-captions-skills[laicut]" --extra-index-url https://lattifai.github.io/pypi/simple/
+
+# Using pip
+pip install "omni-captions-skills[laicut]" --extra-index-url https://lattifai.github.io/pypi/simple/
+```
+
 **Supported languages:** English, Chinese, German, and mixed
 
 **Recommended workflow:** Align before translate (translated text doesn't match original audio)

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/README_zh.md

@@ -64,6 +64,15 @@ Vibe coding 就是不用自己写代码的编程方式。
 
 普通转录的时间戳只是"大概"，LaiCut 使用 [LattifAI](https://lattifai.com/) Lattice-1 模型将文本与音频波形精确匹配，实现**词级精度**。
 
+**安装 LaiCut：**
+```bash
+# 使用 uv（推荐，自动配置包索引）
+uv pip install "omni-captions-skills[laicut]"
+
+# 使用 pip
+pip install "omni-captions-skills[laicut]" --extra-index-url https://lattifai.github.io/pypi/simple/
+```
+
 **支持语言：** 英语、中文、德语及混合
 
 **推荐工作流：** 先对齐再翻译（翻译文本与原始音频不匹配，无法对齐）

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "omni-captions-skills"
-version = "0.1.0"
+version = "0.1.2"
 description = "AI-powered media transcription with Claude Code skills"
 readme = "README.md"
 license = "MIT"
@@ -37,10 +37,10 @@ dev = [
     "ruff>=0.4.0",
 ]
 laicut = [
-    "lattifai[diarization]>=1.3.0",
+    "lattifai[diarization]>=1.3.4",
 ]
 all = [
-    "omnicaptions[dev,laicut]",
+    "omni-captions-skills[dev,laicut]",
 ]
 
 [project.scripts]
@@ -60,3 +60,6 @@ target-version = "py310"
 [tool.ruff.lint]
 select = ["E", "F", "I"]
 ignore = ["E501"]
+
+[tool.uv]
+extra-index-url = ["https://lattifai.github.io/pypi/simple/"]

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/skills/omnicaptions-LaiCut/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: omnicaptions-LaiCut
 description: Use when user needs accurate/precise caption timing, or aligning captions with audio/video using forced alignment. Corrects caption timing to match actual speech. Uses LattifAI Lattice-1 model.
-allowed-tools: Bash(omnicaptions:*)
+allowed-tools: Read, Bash(omnicaptions:*), Bash(lai:*)
 ---
 
 # LaiCut
@@ -27,10 +27,7 @@ LattifAI's audio-text processing toolkit. Currently supports forced alignment, w
 ## Setup
 
 ```bash
-pip install "lattifai-captions[splitting] @ https://github.com/lattifai/omni-captions-skills/raw/main/packages/lattifai_captions-0.1.0.tar.gz"
-pip install https://github.com/lattifai/omni-captions-skills/raw/main/packages/omnicaptions-0.1.0.tar.gz
-pip install https://github.com/lattifai/omni-captions-skills/raw/main/packages/lattifai_core-0.6.1.tar.gz
-pip install "lattifai[alignment] @ https://github.com/lattifai/omni-captions-skills/raw/main/packages/lattifai-1.2.2.tar.gz"
+pip install "omni-captions-skills[laicut]" --extra-index-url https://lattifai.github.io/pypi/simple/
 ```
 
 ## API Key

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/skills/omnicaptions-convert/SKILL.md

@@ -35,8 +35,7 @@ omnicaptions convert transcript.md -o output.vtt
 ## Setup
 
 ```bash
-pip install https://github.com/lattifai/omni-captions-skills/raw/main/packages/lattifai_captions-0.1.0.tar.gz
-pip install https://github.com/lattifai/omni-captions-skills/raw/main/packages/omnicaptions-0.1.0.tar.gz
+pip install omni-captions-skills
 ```
 
 ## Quick Reference

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/skills/omnicaptions-download/SKILL.md

@@ -43,8 +43,7 @@ Only proceed with the download command after user confirms.
 ## Setup
 
 ```bash
-pip install https://github.com/lattifai/omni-captions-skills/raw/main/packages/lattifai_captions-0.1.0.tar.gz
-pip install https://github.com/lattifai/omni-captions-skills/raw/main/packages/omnicaptions-0.1.0.tar.gz
+pip install omni-captions-skills
 ```
 
 ## CLI Usage
@@ -55,6 +54,9 @@ pip install https://github.com/lattifai/omni-captions-skills/raw/main/packages/o
 # Download audio only (default, saves to current directory)
 omnicaptions download "https://www.youtube.com/watch?v=VIDEO_ID"
 
+# Supports bare YouTube video ID (auto-validates via yt-dlp)
+omnicaptions download e882eXLtwkI
+
 # Download video (1080p recommended)
 omnicaptions download "https://youtube.com/watch?v=VIDEO_ID" -q 1080p
 

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/skills/omnicaptions-transcribe/SKILL.md

@@ -60,8 +60,7 @@ omnicaptions transcribe video.mp4
 ## Setup
 
 ```bash
-pip install https://github.com/lattifai/omni-captions-skills/raw/main/packages/lattifai_captions-0.1.0.tar.gz
-pip install https://github.com/lattifai/omni-captions-skills/raw/main/packages/omnicaptions-0.1.0.tar.gz
+pip install omni-captions-skills
 ```
 
 ## API Key

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/skills/omnicaptions-translate/SKILL.md

@@ -41,8 +41,7 @@ Output: `input_Gemini_zh.srt`
 ## Setup
 
 ```bash
-pip install https://github.com/lattifai/omni-captions-skills/raw/main/packages/lattifai_captions-0.1.0.tar.gz
-pip install https://github.com/lattifai/omni-captions-skills/raw/main/packages/omnicaptions-0.1.0.tar.gz
+pip install omni-captions-skills
 ```
 
 ## API Key

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/src/omnicaptions/caption.py

@@ -50,6 +50,45 @@ VIDEO_PLATFORM_PATTERNS = [
     r"x\.com/.*/status",
 ]
 
+# YouTube video ID pattern: 11 alphanumeric chars with hyphens and underscores
+YOUTUBE_VIDEO_ID_RE = re.compile(r"^[a-zA-Z0-9_-]{11}$")
+
+
+def is_youtube_video_id(value: str) -> bool:
+    """Check if a string looks like a bare YouTube video ID."""
+    return bool(YOUTUBE_VIDEO_ID_RE.match(value))
+
+
+def resolve_video_input(url_or_id: str) -> str:
+    """Resolve a video URL or bare YouTube video ID to a full URL.
+
+    If the input is a bare YouTube video ID, validates it via yt-dlp
+    extract_info and converts to a full URL. Otherwise returns as-is.
+
+    Raises:
+        ValueError: If the video ID does not exist or is unavailable.
+    """
+    if not is_youtube_video_id(url_or_id):
+        return url_or_id
+
+    import yt_dlp
+
+    full_url = f"https://www.youtube.com/watch?v={url_or_id}"
+    ydl_opts = {"quiet": True, "no_warnings": True, "skip_download": True}
+    with yt_dlp.YoutubeDL(ydl_opts) as ydl:
+        try:
+            info = ydl.extract_info(full_url, download=False)
+        except yt_dlp.utils.DownloadError as e:
+            raise ValueError(f"YouTube video '{url_or_id}' not found or unavailable: {e}") from e
+        if not info:
+            raise ValueError(f"YouTube video '{url_or_id}' not found or unavailable")
+
+    logging.info(
+        "Resolved video ID '%s' → %s (title: %s)", url_or_id, full_url, info.get("title", "")
+    )
+    return full_url
+
+
 LANGUAGE_NAMES = {
     "zh": "Chinese (Simplified)",
     "zh-TW": "Chinese (Traditional)",
@@ -313,9 +352,9 @@ class GeminiCaption:
         video_path = None
         caption_path = None
 
-        audio_exts = (".webm", ".m4a", ".mp3", ".opus", ".ogg", ".wav")
+        audio_exts = (".mp3", ".m4a", ".mp3", ".opus", ".ogg", ".wav")
         video_exts = (".mp4", ".mkv", ".avi", ".mov", ".flv")
-        caption_exts = (".vtt", ".srt", ".ass")
+        caption_exts = (".vtt", ".srt", ".ass", ".srv3")
 
         for f in output_dir.iterdir():
             # Check if filename starts with video_id
@@ -341,10 +380,11 @@ class GeminiCaption:
             if caption_path:
                 self.logger.info(f"Downloaded caption: {caption_path}")
 
-        # Save metadata to .meta.json for later use (e.g., ASS font scaling)
+        # Save metadata to .meta.json for later use (e.g., ASS font scaling, video_url tracking)
         meta_path = output_dir / f"{video_id}.meta.json"
         meta_data = {
             "video_id": video_id,
+            "video_url": url,
             "title": title,
             "width": video_width,
             "height": video_height,

{omni_captions_skills-0.1.0 → omni_captions_skills-0.1.2}/src/omnicaptions/cli.py

@@ -6,6 +6,13 @@ import subprocess
 import sys
 from pathlib import Path
 
+try:
+    from dotenv import load_dotenv
+
+    load_dotenv()
+except ImportError:
+    pass
+
 from lattifai.caption import Caption, GeminiReader
 
 from .caption import GeminiCaption, GeminiCaptionConfig
@@ -343,8 +350,12 @@ def ensure_api_key(api_key: str | None = None) -> bool:
 
 
 def is_url(path: str) -> bool:
-    """Check if path is a URL."""
-    return path.startswith(("http://", "https://", "youtube.com", "youtu.be"))
+    """Check if path is a URL or a bare YouTube video ID."""
+    from .caption import is_youtube_video_id
+
+    return path.startswith(
+        ("http://", "https://", "youtube.com", "youtu.be")
+    ) or is_youtube_video_id(path)
 
 
 def get_default_output_dir(input_path: str) -> Path:
@@ -363,8 +374,13 @@ def get_stem_from_input(input_path: str) -> str:
     """Extract stem (filename without extension) from input.
 
     For URLs, extract video ID or use 'output'.
+    For bare YouTube video IDs, return the ID directly.
     For files, use the file stem.
     """
+    from .caption import is_youtube_video_id
+
+    if is_youtube_video_id(input_path):
+        return input_path
     if is_url(input_path):
         # Try to extract YouTube video ID
         import re
@@ -461,6 +477,8 @@ def cmd_transcribe(args):
 
 def cmd_download(args):
     """Download audio/video and captions from video platforms."""
+    from .caption import resolve_video_input
+
     config = GeminiCaptionConfig(verbose=args.verbose)
     gc = GeminiCaption(config=config)
 
@@ -469,8 +487,15 @@ def cmd_download(args):
 
     quality = getattr(args, "quality", "audio")
 
+    # Resolve bare video ID to full URL (validates via yt-dlp)
     try:
-        result = gc.download(args.url, output_dir, quality)
+        url = resolve_video_input(args.url)
+    except ValueError as e:
+        print(f"Error: {e}", file=sys.stderr)
+        sys.exit(1)
+
+    try:
+        result = gc.download(url, output_dir, quality)
         if result.video_path:
             print(f"Video: {result.video_path}")
         if result.audio_path:
@@ -691,6 +716,21 @@ def cmd_translate(args):
         sys.exit(1)
 
 
+def _detect_metadata(audio_path: Path) -> dict | None:
+    """Detect metadata from .meta.json in the same directory."""
+    import json
+
+    # Try to find meta.json based on audio filename (e.g., e882eXLtwkI.m4a -> e882eXLtwkI.meta.json)
+    stem = audio_path.stem
+    meta_path = audio_path.parent / f"{stem}.meta.json"
+    if meta_path.exists():
+        try:
+            return json.loads(meta_path.read_text())
+        except Exception:
+            pass
+    return None
+
+
 def cmd_laicut_align(args):
     """Align audio with caption using LattifAI forced alignment."""
     # Get API key
@@ -709,6 +749,9 @@ def cmd_laicut_align(args):
     audio_path = Path(args.audio)
     caption_path = Path(args.caption)
 
+    # Auto-detect metadata from .meta.json
+    metadata = _detect_metadata(audio_path)
+
     if not audio_path.exists():
         print(f"Error: Audio file not found: {audio_path}", file=sys.stderr)
         sys.exit(1)
@@ -718,7 +761,7 @@ def cmd_laicut_align(args):
         sys.exit(1)
 
     from lattifai.client import LattifAI
-    from lattifai.config import AlignmentConfig, CaptionConfig, ClientConfig
+    from lattifai.config import AlignmentConfig, CaptionConfig, ClientConfig, EventConfig
 
     # Determine output path and format
     if args.output:
@@ -746,12 +789,16 @@ def cmd_laicut_align(args):
                 split_sentence=getattr(args, "split_sentence", False),
                 word_level=word_level,
             ),
+            event_config=EventConfig(
+                enabled=True,
+            ),
         )
 
         client.alignment(
             input_media=str(audio_path),
             input_caption=str(caption_path),
             output_caption_path=str(output_path),
+            metadata=metadata,
         )
         print(f"LaiCut aligned: {output_path}")
     except Exception as e:
@@ -793,7 +840,7 @@ def main():
     p_download = subparsers.add_parser(
         "download", help="Download audio/video and captions from URL"
     )
-    p_download.add_argument("url", help="Video URL (YouTube, Bilibili, etc.)")
+    p_download.add_argument("url", help="Video URL or YouTube video ID (e.g. e882eXLtwkI)")
     p_download.add_argument("-o", "--output", help="Output directory (default: current)")
     p_download.add_argument(
         "-q",