diff --git a/frontend/src/i18n/locales/en-US.ts b/frontend/src/i18n/locales/en-US.ts index 4b369fb1..7534742f 100755 --- a/frontend/src/i18n/locales/en-US.ts +++ b/frontend/src/i18n/locales/en-US.ts @@ -1516,7 +1516,7 @@ export default { faq: 'FAQ Settings', graph: 'Knowledge Graph', multimodal: 'Image Processing', - asr: 'Audio Processing', + asr: 'Audio & Video', datasource: 'Data Sources', share: 'Sharing' }, @@ -1724,12 +1724,13 @@ export default { description: 'Configure image content understanding for parsing and retrieving non-text content like images', }, asr: { - title: 'Audio Processing Configuration', - description: 'Configure speech recognition to upload audio files (mp3, wav, m4a, flac, ogg) and automatically transcribe to text', - label: 'Enable Speech Recognition', - desc: 'When enabled, audio files can be uploaded to the knowledge base and automatically transcribed to text', + title: 'Audio & Video Speech Recognition', + description: + 'Configure ASR (speech-to-text). When enabled, you can upload audio files or videos; audio is transcribed directly, and video audio tracks are extracted and transcribed (e.g. mp3, wav, m4a, flac, ogg, mp4, mov, …).', + label: 'Enable audio & video speech recognition', + desc: 'When enabled, audio or video can be uploaded to the knowledge base; speech is transcribed to text for parsing and retrieval.', modelLabel: 'ASR Model', - modelDescription: 'Speech-to-text model for audio transcription (e.g. OpenAI Whisper)', + modelDescription: 'Speech-to-text model for audio and video soundtracks (e.g. OpenAI Whisper)', modelPlaceholder: 'Select an ASR model', }, advanced: { @@ -3179,7 +3180,7 @@ export default { fileTypeText: 'Plain Text', fileTypeJson: 'JSON Files', fileTypeImage: 'Images', - fileTypeAudio: 'Audio Files', + fileTypeAudiovisual: 'Audio & Video', engines: { builtin: { name: 'Built-in', diff --git a/frontend/src/i18n/locales/ko-KR.ts b/frontend/src/i18n/locales/ko-KR.ts index b22ebf5c..39833a21 100755 --- a/frontend/src/i18n/locales/ko-KR.ts +++ b/frontend/src/i18n/locales/ko-KR.ts @@ -2037,7 +2037,7 @@ export default { faq: "FAQ 설정", graph: "지식 그래프", multimodal: "이미지 처리", - asr: "오디오 처리", + asr: "오디오·영상", share: "공유관리", storage: "스토리지 엔진", datasource: "데이터 소스", @@ -2256,12 +2256,13 @@ export default { description: "이미지 콘텐츠 이해 기능을 설정하여 이미지 등 비텍스트 콘텐츠의 파싱 및 검색을 지원합니다", }, asr: { - title: "오디오 처리 설정", - description: "음성 인식 기능을 설정하여 오디오 파일(mp3, wav, m4a, flac, ogg)을 업로드하고 자동으로 텍스트로 전사합니다", - label: "음성 인식 활성화", - desc: "활성화하면 오디오 파일을 지식베이스에 업로드하고 자동으로 텍스트로 전사할 수 있습니다", + title: "오디오·영상 음성 인식", + description: + "ASR(음성-텍스트)을 설정합니다. 활성화 시 오디오 또는 영상을 업로드할 수 있으며, 오디오는 그대로 전사하고 영상은 음성 트랙을 추출해 텍스트로 전사합니다(mp3, wav, m4a, flac, ogg, mp4, mov 등).", + label: "오디오·영상 음성 인식 사용", + desc: "활성화하면 오디오 또는 영상을 지식베이스에 업로드할 수 있으며, 음성이 텍스트로 전사되어 파싱·검색에 사용됩니다", modelLabel: "ASR 모델", - modelDescription: "오디오 전사를 위한 음성 인식 모델 (예: OpenAI Whisper)", + modelDescription: "오디오·영상의 음성을 텍스트로 전사하는 모델 (예: OpenAI Whisper)", modelPlaceholder: "ASR 모델을 선택해주세요", }, advanced: { @@ -3234,7 +3235,7 @@ export default { fileTypeText: '일반 텍스트', fileTypeJson: 'JSON 파일', fileTypeImage: '이미지', - fileTypeAudio: '오디오 파일', + fileTypeAudiovisual: '오디오·영상', engines: { builtin: { name: '내장', diff --git a/frontend/src/i18n/locales/ru-RU.ts b/frontend/src/i18n/locales/ru-RU.ts index 02a3ae17..45a775ad 100755 --- a/frontend/src/i18n/locales/ru-RU.ts +++ b/frontend/src/i18n/locales/ru-RU.ts @@ -1717,7 +1717,7 @@ export default { faq: 'FAQ настройки', graph: 'Граф знаний', multimodal: 'Обработка изображений', - asr: 'Обработка аудио', + asr: 'Аудио и видео', storage: 'Storage Engine', datasource: 'Источники данных', share: 'Sharing' @@ -1916,12 +1916,13 @@ export default { description: 'Настройте понимание изображений для парсинга и поиска нетекстового контента', }, asr: { - title: 'Обработка аудио', - description: 'Настройте распознавание речи для загрузки аудиофайлов (mp3, wav, m4a, flac, ogg) и автоматической транскрибации в текст', - label: 'Включить распознавание речи', - desc: 'При включении аудиофайлы могут быть загружены в базу знаний и автоматически транскрибированы в текст', + title: 'Распознавание речи для аудио и видео', + description: + 'Настройте ASR (речь → текст). При включении можно загружать аудио или видео: аудио транскрибируется целиком, из видео извлекается дорожка и транскрибируется (mp3, wav, m4a, flac, ogg, mp4, mov и др.).', + label: 'Включить распознавание для аудио и видео', + desc: 'При включении в базу можно загружать аудио или видео; речь транскрибируется в текст для парсинга и поиска', modelLabel: 'Модель ASR', - modelDescription: 'Модель распознавания речи для транскрибации аудио (например, OpenAI Whisper)', + modelDescription: 'Модель распознавания речи для транскрибации аудио и звуковых дорожек видео (например, OpenAI Whisper)', modelPlaceholder: 'Выберите модель ASR', }, advanced: { @@ -2849,7 +2850,7 @@ export default { fileTypeText: 'Текстовые файлы', fileTypeJson: 'Файлы JSON', fileTypeImage: 'Изображения', - fileTypeAudio: 'Аудиофайлы', + fileTypeAudiovisual: 'Аудио и видео', engines: { builtin: { name: 'Встроенный', diff --git a/frontend/src/i18n/locales/zh-CN.ts b/frontend/src/i18n/locales/zh-CN.ts index 3411e03f..bc141af7 100755 --- a/frontend/src/i18n/locales/zh-CN.ts +++ b/frontend/src/i18n/locales/zh-CN.ts @@ -2023,7 +2023,7 @@ export default { faq: "FAQ 设置", graph: "知识图谱", multimodal: "图像处理", - asr: "音频处理", + asr: "音视频", datasource: "数据源", share: "共享管理", }, @@ -2228,12 +2228,12 @@ export default { description: "配置图像内容理解能力,启用后支持图片等非文本内容的解析和检索", }, asr: { - title: "音频处理配置", - description: "配置语音识别功能,启用后支持上传音频文件(mp3、wav、m4a、flac、ogg)并自动转录为文本", - label: "启用语音识别", - desc: "启用后可以上传音频文件到知识库,系统将自动转录为文本进行处理", + title: "音视频语音识别", + description: "配置语音识别(ASR),启用后支持上传音频及视频文件;音频将整段转写,视频将提取音轨并转写为文本(常见格式:mp3、wav、m4a、flac、ogg、mp4、mov 等)", + label: "启用音视频语音识别", + desc: "启用后可上传音频或视频到知识库,系统自动转写音轨/语音为文本并参与解析与检索", modelLabel: "ASR 模型", - modelDescription: "用于语音转文本的语音识别模型(如 OpenAI Whisper)", + modelDescription: "用于音频与视频中语音转文本的识别模型(如 OpenAI Whisper)", modelPlaceholder: "请选择 ASR 模型", }, advanced: { @@ -3177,7 +3177,7 @@ export default { fileTypeText: "纯文本", fileTypeJson: "JSON 文件", fileTypeImage: "图片", - fileTypeAudio: "音频文件", + fileTypeAudiovisual: "音视频", engines: { builtin: { name: "内置", diff --git a/frontend/src/views/knowledge/KnowledgeBase.vue b/frontend/src/views/knowledge/KnowledgeBase.vue index 7748949e..c6ed91ee 100644 --- a/frontend/src/views/knowledge/KnowledgeBase.vue +++ b/frontend/src/views/knowledge/KnowledgeBase.vue @@ -1030,11 +1030,38 @@ const handleDocumentUpload = async (event: Event) => { return; } + const vlmEnabled = kbInfo.value?.vlm_config?.enabled || false; + const asrEnabled = kbInfo.value?.asr_config?.enabled || false; const dynamicTypes = supportedFileTypes.value.size > 0 ? supportedFileTypes.value : undefined const validFiles: File[] = []; let skippedCount = 0; + let imageFilteredCount = 0; + let videoFilteredCount = 0; + let audioFilteredCount = 0; + for (let i = 0; i < files.length; i++) { const file = files[i]; + const fileExt = file.name.substring(file.name.lastIndexOf('.') + 1).toLowerCase(); + const imageTypes = ['jpg', 'jpeg', 'png', 'gif', 'bmp', 'webp']; + const videoTypes = ['mp4', 'mov', 'avi', 'mkv', 'webm', 'wmv', 'flv']; + const audioTypes = ['mp3', 'wav', 'm4a', 'flac', 'ogg']; + + if (!vlmEnabled) { + if (imageTypes.includes(fileExt)) { + imageFilteredCount++; + continue; + } + if (videoTypes.includes(fileExt)) { + videoFilteredCount++; + continue; + } + } + + if (!asrEnabled && audioTypes.includes(fileExt)) { + audioFilteredCount++; + continue; + } + if (!kbFileTypeVerification(file, files.length > 1, dynamicTypes)) { validFiles.push(file); } else { @@ -1042,6 +1069,16 @@ const handleDocumentUpload = async (event: Event) => { } } + if (imageFilteredCount > 0) { + MessagePlugin.warning(t('knowledgeBase.imagesFilteredNoVLM', { count: imageFilteredCount })); + } + if (videoFilteredCount > 0) { + MessagePlugin.warning(t('knowledgeBase.videosFilteredNoVLM', { count: videoFilteredCount })); + } + if (audioFilteredCount > 0) { + MessagePlugin.warning(t('knowledgeBase.audiosFilteredNoASR', { count: audioFilteredCount })); + } + if (validFiles.length === 0) { if (skippedCount > 0) { MessagePlugin.warning(t('knowledgeBase.allFilesSkippedNoEngine')); @@ -1176,6 +1213,16 @@ const handleFolderUpload = async (event: Event) => { } } + if (imageFilteredCount > 0) { + MessagePlugin.warning(t('knowledgeBase.imagesFilteredNoVLM', { count: imageFilteredCount })); + } + if (videoFilteredCount > 0) { + MessagePlugin.warning(t('knowledgeBase.videosFilteredNoVLM', { count: videoFilteredCount })); + } + if (audioFilteredCount > 0) { + MessagePlugin.warning(t('knowledgeBase.audiosFilteredNoASR', { count: audioFilteredCount })); + } + if (validFiles.length === 0) { MessagePlugin.warning(t('knowledgeBase.noValidFilesInFolder', { total: files.length })); if (input) input.value = ''; diff --git a/frontend/src/views/knowledge/KnowledgeBaseEditorModal.vue b/frontend/src/views/knowledge/KnowledgeBaseEditorModal.vue index f11d833c..e295bcf2 100644 --- a/frontend/src/views/knowledge/KnowledgeBaseEditorModal.vue +++ b/frontend/src/views/knowledge/KnowledgeBaseEditorModal.vue @@ -202,7 +202,7 @@ - +