starmen1265/kb-skills
Archived
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
433 Commits 3 Branches 4 Tags
20a80fcb04f07c97d1b2c509229faa33d9cf44b5
Commit Graph

9 Commits

Author SHA1 Message Date
starmen1265 759503f48a feat(13-05): 实现中文句边界检测指令 
更新 SEGMENTATION_PROMPT，增强句边界检测：
- 列举所有中文句末标点（。？！……）
- 强调「自然停顿处」和「语义停顿」
- 说明「不仅依赖标点符号」

添加测试验证句边界检测效果：
- test_chinese_sentence_boundary_punctuations
- test_prompt_emphasizes_semantic_pause
- test_prompt_not_only_punctuation
- test_sentence_boundary_in_segmentation

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-22 00:25:40 +08:00
starmen1265 00319f523f feat(13-01): implement video type detection and Pydantic models 
- 创建 SemanticSegmentationEngine 类
- 实现 VideoSegment 和 SegmentationResult Pydantic 模型
  - VideoSegment: title(可选), content(必填), start_index, end_index
  - SegmentationResult: video_type, segments[], actual_segment_count
  - 添加标题长度验证器（5-15字）
- 实现 _infer_video_type() 方法
  - 从元数据推断视频类型（知识类/故事类/访谈类/新闻类/vlog）
  - 支持从 title, category, tags 推断
  - 默认类型为"知识类"
- 实现 _build_prompt() 方法
  - 包含视频类型信息
  - 包含完整的分段要求（数量控制、句边界保护、标题生成等）
  - 包含 JSON 输出格式规范
- 添加 extract_json_from_llm_response() 工具函数
- 添加 AIClientProtocol 协议定义
- 添加 deferred-items.md 记录超出范围的测试

所有计划 13-01 测试通过（10个测试）

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-22 00:25:21 +08:00
starmen1265 8c2b3e079a feat(13-04): 实现主题标题生成功能 
实现 VideoSegment.title 字段和验证，Prompt 中的标题生成指令。

功能：
- VideoSegment.title: Optional[str] 字段
- @field_validator 验证标题长度（不超过 15 字）
- SEGMENTATION_PROMPT 包含标题生成指令
- LLM 自主判断是否需要标题（主题/章节有标题，论点/观点无标题）
- 测试验证标题生成和省略

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-22 00:24:58 +08:00
starmen1265 a2ca45523b fix(srt): add punctuation requirement to ASR correction prompt 
- Update CORRECTION_PROMPT to explicitly require adding punctuation
- Fix missing punctuation in thunderstorm and lightning sections

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-21 23:20:07 +08:00
starmen1265 4d2645876b fix(srt): add disable_reasoning parameter for ASR correction and include title in archive filename
PR Test / lint (pull_request) Successful in 1s
Details 
- SimpleAIClient: Add disable_reasoning parameter to control AI thinking mode
- ASRCorrector: Use disable_reasoning=True to get direct correction results
- ArchiveProcessor: Extract title from frontmatter and include in filename

File naming format: {YYYYMMDD}_{content_id}_{title}_analysis.md

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-21 22:19:14 +08:00
starmen1265 812f016b1b chore(12-01): update srt module exports 
- Add SemanticSegmenter and ASRCorrector to __all__
- Maintain lazy loading for optional dependencies

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-19 23:39:20 +08:00
starmen1265 5e00632f68 feat(12-01): implement ASRCorrector for speech recognition error correction 
- Use LLM to correct obvious ASR errors (homophones, missing words)
- Batch processing for long text (5000 chars per batch)
- Fallback to original text when AI unavailable
- Simple sentence-based text splitting

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-19 23:35:53 +08:00
starmen1265 a350693863 feat(12-01): implement SemanticSegmenter for subtitle segmentation 
- Merge short segments into paragraphs
- Split on time gaps (>3 seconds by default)
- Enforce max paragraph length (500 chars by default)
- Simple rule-based approach (no LLM needed)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-19 23:34:18 +08:00
starmen1265 c259e0ac2f feat(12-01): implement SRT parser module 
- Add Segment dataclass for subtitle segments
- Add parse_srt() function to parse SRT content string
- Add parse_srt_file() function to parse from file path
- Support standard SRT format with timestamps
- Handle special characters and multi-line subtitles

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-19 23:32:57 +08:00