feat(annotation): 支持音频和视频数据类型的标注任务

- 添加了音频和视频数据类型常量定义
- 实现了音频和视频标注模板的内置配置
- 扩展前端组件以支持按数据类型过滤标注模板
- 重构后端编辑器服务以处理音频和视频任务构建
- 更新数据库初始化脚本包含音频和视频标注模板
- 添加音频和视频数据类型的预览URL映射逻辑

runtime/datamate-python/app/module/annotation/service/builtin_templates.py

@@ -13,7 +13,11 @@ from app.module.annotation.utils.config_validator import LabelStudioConfigValida
 logger = get_logger(__name__)
 
 DATA_TYPE_IMAGE = "image"
+DATA_TYPE_AUDIO = "audio"
+DATA_TYPE_VIDEO = "video"
 CATEGORY_COMPUTER_VISION = "computer-vision"
+CATEGORY_AUDIO_SPEECH = "audio-speech"
+CATEGORY_VIDEO = "video"
 STYLE_HORIZONTAL = "horizontal"
 VERSION_DEFAULT = "1.0.0"
 
@@ -51,6 +55,105 @@ SEMANTIC_SEGMENTATION_POLYGON_LABEL_CONFIG = """<View>
   </PolygonLabels>
 </View>"""
 
+ASR_SEGMENTS_LABEL_CONFIG = """<View>
+  <Labels name=\"labels\" toName=\"audio\">
+    <Label value=\"Speech\" />
+    <Label value=\"Noise\" />
+  </Labels>
+  <Audio name=\"audio\" value=\"$audio\"/>
+  <TextArea name=\"transcription\" toName=\"audio\"
+            rows=\"2\" editable=\"true\"
+            perRegion=\"true\" required=\"true\" />
+</View>"""
+
+ASR_LABEL_CONFIG = """<View>
+  <Audio name=\"audio\" value=\"$audio\" zoom=\"true\" hotkey=\"ctrl+enter\" />
+  <Header value=\"转录音频内容\" />
+  <TextArea name=\"transcription\" toName=\"audio\"
+            rows=\"4\" editable=\"true\" maxSubmissions=\"1\" />
+</View>"""
+
+CONVERSATION_ANALYSIS_LABEL_CONFIG = """<View>
+  <Audio name=\"audio\" value=\"$audio\" hotkey=\"space\" sync=\"text\"/>
+  <Header value=\"对话记录\"/>
+  <Paragraphs audioUrl=\"$audio\" sync=\"audio\" name=\"text\" value=\"$text\"
+              layout=\"dialogue\" textKey=\"text\" nameKey=\"author\"
+              granularity=\"paragraph\" contextscroll=\"true\" />
+  <View style=\"position: sticky\">
+    <Header value=\"情感标签\"/>
+    <ParagraphLabels name=\"label\" toName=\"text\">
+      <Label value=\"Positive\" background=\"#00ff00\"/>
+      <Label value=\"Negative\" background=\"#ff0000\"/>
+    </ParagraphLabels>
+  </View>
+</View>"""
+
+INTENT_CLASSIFICATION_LABEL_CONFIG = """<View>
+  <Labels name=\"labels\" toName=\"audio\">
+    <Label value=\"Segment\" />
+  </Labels>
+  <Audio name=\"audio\" value=\"$audio\"/>
+  <Choices name=\"intent\" toName=\"audio\" perRegion=\"true\" required=\"true\">
+    <Choice value=\"Question\" />
+    <Choice value=\"Request\" />
+    <Choice value=\"Satisfied\" />
+    <Choice value=\"Interested\" />
+    <Choice value=\"Unsatisfied\" />
+  </Choices>
+</View>"""
+
+SIGNAL_QUALITY_LABEL_CONFIG = """<View>
+  <Rating name=\"rating\" toName=\"audio\" maxRating=\"10\" icon=\"star\" size=\"medium\" />
+  <Audio name=\"audio\" value=\"$audio\"/>
+</View>"""
+
+SOUND_EVENT_DETECTION_LABEL_CONFIG = """<View>
+  <Labels name=\"label\" toName=\"audio\" zoom=\"true\" hotkey=\"ctrl+enter\">
+    <Label value=\"Event A\" background=\"red\"/>
+    <Label value=\"Event B\" background=\"green\"/>
+  </Labels>
+  <Audio name=\"audio\" value=\"$audio\"/>
+</View>"""
+
+SPEAKER_SEGMENTATION_LABEL_CONFIG = """<View>
+  <Labels name=\"label\" toName=\"audio\" zoom=\"true\" hotkey=\"ctrl+enter\">
+    <Label value=\"Speaker one\" background=\"#00FF00\"/>
+    <Label value=\"Speaker two\" background=\"#12ad59\"/>
+  </Labels>
+  <Audio name=\"audio\" value=\"$audio\" />
+</View>"""
+
+VIDEO_CLASSIFICATION_LABEL_CONFIG = """<View>
+  <Video name=\"video\" value=\"$video\"/>
+  <Choices name=\"choice\" toName=\"video\" showInline=\"true\">
+    <Choice value=\"Sports\" />
+    <Choice value=\"News\" />
+    <Choice value=\"Entertainment\" />
+    <Choice value=\"Education\" />
+  </Choices>
+</View>"""
+
+VIDEO_OBJECT_TRACKING_LABEL_CONFIG = """<View>
+  <Labels name=\"videoLabels\" toName=\"video\" allowEmpty=\"true\">
+    <Label value=\"Man\" background=\"blue\"/>
+    <Label value=\"Woman\" background=\"red\"/>
+    <Label value=\"Other\" background=\"green\"/>
+  </Labels>
+  <Video name=\"video\" value=\"$video\" framerate=\"25.0\"/>
+  <VideoRectangle name=\"box\" toName=\"video\" />
+</View>"""
+
+VIDEO_TIMELINE_SEGMENTATION_LABEL_CONFIG = """<View>
+  <Header value=\"视频时间线分割\"/>
+  <Video name=\"video\" value=\"$video_url\" sync=\"audio\"/>
+  <Labels name=\"tricks\" toName=\"audio\" choice=\"multiple\">
+    <Label value=\"Intro\" background=\"#358EF3\"/>
+    <Label value=\"Content\" background=\"#1BB500\"/>
+    <Label value=\"Outro\" background=\"#FFA91D\"/>
+  </Labels>
+  <Audio name=\"audio\" value=\"$video_url\" sync=\"video\" speed=\"false\"/>
+</View>"""
+
 
 @dataclass(frozen=True)
 class BuiltInTemplateDefinition:
@@ -122,6 +225,122 @@ BUILT_IN_TEMPLATES: List[BuiltInTemplateDefinition] = [
         category=CATEGORY_COMPUTER_VISION,
         version=VERSION_DEFAULT,
     ),
+    BuiltInTemplateDefinition(
+        id="tpl-asr-segments-001",
+        name="语音识别（分段）",
+        description=(
+            "对音频进行语音活动分段并转录文本，适用于呼叫中心转录、会议记录、播客转录等场景。"
+            "关联模型：Whisper、Wav2Vec2、DeepSpeech"
+        ),
+        data_type=DATA_TYPE_AUDIO,
+        labeling_type="asr-segments",
+        label_config=ASR_SEGMENTS_LABEL_CONFIG,
+        style=STYLE_HORIZONTAL,
+        category=CATEGORY_AUDIO_SPEECH,
+        version=VERSION_DEFAULT,
+    ),
+    BuiltInTemplateDefinition(
+        id="tpl-asr-001",
+        name="语音识别",
+        description=(
+            "转录音频内容，适用于播客转录、会议记录、客服通话、字幕生成等场景。"
+            "关联模型：Whisper、Wav2Vec、DeepSpeech"
+        ),
+        data_type=DATA_TYPE_AUDIO,
+        labeling_type="asr",
+        label_config=ASR_LABEL_CONFIG,
+        style=STYLE_HORIZONTAL,
+        category=CATEGORY_AUDIO_SPEECH,
+        version=VERSION_DEFAULT,
+    ),
+    BuiltInTemplateDefinition(
+        id="tpl-conversation-analysis-001",
+        name="对话分析",
+        description="分析对话语句并标注事实和情感方面，适用于呼叫中心质检、客服分析、会议分析等场景",
+        data_type=DATA_TYPE_AUDIO,
+        labeling_type="conversation-analysis",
+        label_config=CONVERSATION_ANALYSIS_LABEL_CONFIG,
+        style=STYLE_HORIZONTAL,
+        category=CATEGORY_AUDIO_SPEECH,
+        version=VERSION_DEFAULT,
+    ),
+    BuiltInTemplateDefinition(
+        id="tpl-intent-classification-001",
+        name="意图分类",
+        description="进行语音活动分段并选择语音意图，适用于语音助手、智能音箱、IVR系统等场景",
+        data_type=DATA_TYPE_AUDIO,
+        labeling_type="intent-classification",
+        label_config=INTENT_CLASSIFICATION_LABEL_CONFIG,
+        style=STYLE_HORIZONTAL,
+        category=CATEGORY_AUDIO_SPEECH,
+        version=VERSION_DEFAULT,
+    ),
+    BuiltInTemplateDefinition(
+        id="tpl-signal-quality-001",
+        name="信号质量检测",
+        description="评估音频信号质量，适用于电信、呼叫中心质检、音频制作、VoIP质量评估等场景",
+        data_type=DATA_TYPE_AUDIO,
+        labeling_type="signal-quality",
+        label_config=SIGNAL_QUALITY_LABEL_CONFIG,
+        style=STYLE_HORIZONTAL,
+        category=CATEGORY_AUDIO_SPEECH,
+        version=VERSION_DEFAULT,
+    ),
+    BuiltInTemplateDefinition(
+        id="tpl-sound-event-001",
+        name="声音事件检测",
+        description="选择音频片段并分类声音事件，适用于安防监控、智慧城市、环境监测、工业监测等场景",
+        data_type=DATA_TYPE_AUDIO,
+        labeling_type="sound-event-detection",
+        label_config=SOUND_EVENT_DETECTION_LABEL_CONFIG,
+        style=STYLE_HORIZONTAL,
+        category=CATEGORY_AUDIO_SPEECH,
+        version=VERSION_DEFAULT,
+    ),
+    BuiltInTemplateDefinition(
+        id="tpl-speaker-segmentation-001",
+        name="说话人分割",
+        description="执行说话人分割/话者分离任务，适用于会议转录、播客制作、呼叫中心分析等场景",
+        data_type=DATA_TYPE_AUDIO,
+        labeling_type="speaker-segmentation",
+        label_config=SPEAKER_SEGMENTATION_LABEL_CONFIG,
+        style=STYLE_HORIZONTAL,
+        category=CATEGORY_AUDIO_SPEECH,
+        version=VERSION_DEFAULT,
+    ),
+    BuiltInTemplateDefinition(
+        id="tpl-video-classification-001",
+        name="视频分类",
+        description="对视频进行整体分类，适用于内容审核、媒体分析、质检等场景",
+        data_type=DATA_TYPE_VIDEO,
+        labeling_type="video-classification",
+        label_config=VIDEO_CLASSIFICATION_LABEL_CONFIG,
+        style=STYLE_HORIZONTAL,
+        category=CATEGORY_VIDEO,
+        version=VERSION_DEFAULT,
+    ),
+    BuiltInTemplateDefinition(
+        id="tpl-video-object-tracking-001",
+        name="视频目标追踪",
+        description="在视频中追踪目标对象，适用于安防监控、交通分析、行为分析等场景",
+        data_type=DATA_TYPE_VIDEO,
+        labeling_type="video-object-tracking",
+        label_config=VIDEO_OBJECT_TRACKING_LABEL_CONFIG,
+        style=STYLE_HORIZONTAL,
+        category=CATEGORY_VIDEO,
+        version=VERSION_DEFAULT,
+    ),
+    BuiltInTemplateDefinition(
+        id="tpl-video-timeline-segmentation-001",
+        name="视频时间线分割",
+        description="对视频时间线进行分段标注，适用于视频剪辑、内容索引等场景",
+        data_type=DATA_TYPE_VIDEO,
+        labeling_type="video-timeline-segmentation",
+        label_config=VIDEO_TIMELINE_SEGMENTATION_LABEL_CONFIG,
+        style=STYLE_HORIZONTAL,
+        category=CATEGORY_VIDEO,
+        version=VERSION_DEFAULT,
+    ),
 ]
 
 assert len({template.id for template in BUILT_IN_TEMPLATES}) == len(BUILT_IN_TEMPLATES), (