feat: 完善群聊日报语音转写缓存

支持空内容的微信语音消息并兼容历史缓存转写结果将系统通知标记为微信系统消息，排除活跃成员与发言排行
2026-08-17 11:37:06 +08:00 · 2026-08-11 11:08:55 +08:00
parent 68f0c0b5a3
commit 6192e7cd35
16 changed files with 240 additions and 42 deletions
@@ -7,7 +7,8 @@ import {
  GroupReportExportResult,
  GroupReportMetadata,
  ReportHeat,
-  ReportSectionMeta
+  ReportSectionMeta,
+  selectHeroParticipantNames
 } from '../shared/group-report'
 import { resolveMd5, getGroupSnapshot } from './services/chat-service'
 import { imageInsightService } from './services/image-insight-service'
@@ -186,11 +187,11 @@ const renderReportHtml = async (request: GroupReportExportRequest): Promise<stri
  )
  const avatar = (name: string): string => avatars.get(name) || fallbackAvatar(name)

-  const heroNames = metadata.heroParticipants.slice(0, 4)
-  while (heroNames.length < 4) heroNames.push(metadata.groupName)
+  const heroNames = selectHeroParticipantNames(metadata.heroParticipants)
  const heroAvatars = heroNames
    .map((name) => `<img src="${avatar(name)}" alt="${escapeHtml(name)}">`)
    .join('')
+  const heroAvatarClass = heroNames.length ? `avatar-count-${heroNames.length}` : 'empty-section'

  const topicCards = report.topics
    .map(
@@ -218,10 +219,18 @@ const renderReportHtml = async (request: GroupReportExportRequest): Promise<stri
                  if (insight) {
                    // insight 不含 imageUrl,需要按 md5/datName 重新拿;这里通过 ImageDecryptService 间接获取
                    // 走 ImageDecryptService.findImageFile + decryptImageToBase64
-                    const decryptService = (globalThis as { __imageDecrypt?: { findImageFile: (md5?: string, dat?: string) => string | null; decryptImageToBase64: (p: string) => string | null } }).__imageDecrypt
+                    const decryptService = (
+                      globalThis as {
+                        __imageDecrypt?: {
+                          findImageFile: (md5?: string, dat?: string) => string | null
+                          decryptImageToBase64: (p: string) => string | null
+                        }
+                      }
+                    ).__imageDecrypt
                    if (decryptService) {
                      const filePath = decryptService.findImageFile(insight.md5, insight.datName)
-                      if (filePath) imageUrl = decryptService.decryptImageToBase64(filePath) || undefined
+                      if (filePath)
+                        imageUrl = decryptService.decryptImageToBase64(filePath) || undefined
                    }
                  }
                }
@@ -451,7 +460,9 @@ const renderReportHtml = async (request: GroupReportExportRequest): Promise<stri
    DATE_RANGE: escapeHtml(metadata.dateRange),
    RECORD_NOTE: escapeHtml(metadata.recordNote),
    // v1 模板使用的 OVERVIEW(经典版以概览段落呈现)
-    OVERVIEW: escapeHtml(report.overview || report.hero?.summary || '基于已读取聊天记录生成的群聊日报'),
+    OVERVIEW: escapeHtml(
+      report.overview || report.hero?.summary || '基于已读取聊天记录生成的群聊日报'
+    ),
    // v2 模板使用的 hero-*
    HERO_HEADLINE: escapeHtml(report.hero?.headline || '今日群聊速览'),
    HERO_SUMMARY: escapeHtml(report.hero?.summary || report.overview),
@@ -462,6 +473,7 @@ const renderReportHtml = async (request: GroupReportExportRequest): Promise<stri
    HERO_TAKEAWAY_EMPTY_CLASS: report.hero?.keyTakeaway ? '' : 'empty-section',
    HERO_PENDING_EMPTY_CLASS: report.hero?.pendingNote ? '' : 'empty-section',
    HERO_AVATARS: heroAvatars,
+    HERO_AVATAR_CLASS: heroAvatarClass,
    MESSAGE_COUNT: String(summaryStats.messageCount),
    ACTIVE_USERS: String(summaryStats.activeUsers),
    TIME_SPAN: escapeHtml(metadata.timeSpan || ''),
@@ -473,13 +485,21 @@ const renderReportHtml = async (request: GroupReportExportRequest): Promise<stri
    RESOURCES_EMPTY_CLASS: sectionClass(request, 'resources', report.resources.length > 0),
    RESOURCE_ITEMS: resourceItems,
    RESOURCES_MORE_NOTE: overflowNote(request, 'resources'),
-    MESSAGES_EMPTY_CLASS: sectionClass(request, 'importantMessages', report.importantMessages.length > 0),
+    MESSAGES_EMPTY_CLASS: sectionClass(
+      request,
+      'importantMessages',
+      report.importantMessages.length > 0
+    ),
    IMPORTANT_MESSAGES: importantMessages,
    MESSAGES_MORE_NOTE: overflowNote(request, 'importantMessages'),
    QUOTES_EMPTY_CLASS: sectionClass(request, 'moments', report.quotes.length > 0),
    QUOTE_BLOCKS: quoteBlocks,
    QUOTES_MORE_NOTE: overflowNote(request, 'moments'),
-    ACTIONS_EMPTY_CLASS: sectionClass(request, 'actions', report.todos.length + report.unresolved.length > 0),
+    ACTIONS_EMPTY_CLASS: sectionClass(
+      request,
+      'actions',
+      report.todos.length + report.unresolved.length > 0
+    ),
    TODO_EMPTY_CLASS: report.todos.length ? '' : 'empty-section',
    TODO_CARDS: todoCards,
    UNRESOLVED_EMPTY_CLASS: report.unresolved?.length ? '' : 'empty-section',
@@ -511,7 +531,11 @@ const renderReportHtml = async (request: GroupReportExportRequest): Promise<stri
    VOICE_EMPTY_CLASS: sectionClass(request, 'voices', report.media?.voiceHighlights?.length > 0),
    VOICE_CARDS: voiceCards,
    VOICE_MORE_NOTE: overflowNote(request, 'voices'),
-    VOICE_RANK_EMPTY_CLASS: sectionClass(request, 'voices', report.analytics.voiceLeaderboard?.length > 0),
+    VOICE_RANK_EMPTY_CLASS: sectionClass(
+      request,
+      'voices',
+      report.analytics.voiceLeaderboard?.length > 0
+    ),
    VOICE_RANK_CARDS: voiceRankCards,
    BADGES_EMPTY_CLASS: sectionClass(request, 'badges', report.media?.funBadges?.length > 0),
    BADGE_CARDS: badgeCards,
@@ -1114,6 +1114,10 @@ app.whenReady().then(async () => {
    return voiceRecognition.recognize(reference)
  })

+  ipcMain.handle('voice:getTranscriptSnapshot', (_, reference: VoiceMessageReference) => {
+    return voiceRecognition?.getTranscriptSnapshot(reference) || { state: 'pending' as const }
+  })
+
  ipcMain.handle('voice:getBatchPreflight', (_, request: VoiceBatchRequest) => {
    if (!voiceBatchService) throw new Error('Voice recognition is not initialized')
    return voiceBatchService.preflight(request)
@@ -123,6 +123,9 @@ export type ParsedContent =
  | UnknownContent

 export function parseMessageContent(content: string, messageType: number): ParsedContent {
+  // Voice rows may keep their binary payload outside msgContent, so an empty
+  // content string is still a valid voice message.
+  if (messageType === 34) return { type: 'voice' }
  if (!content || typeof content !== 'string') {
    return { type: 'unknown', raw: content || '' }
  }
@@ -132,8 +135,6 @@ export function parseMessageContent(content: string, messageType: number): Parse
  switch (messageType) {
    case 1:
      return { type: 'text', content: normalized }
-    case 34:
-      return { type: 'voice' }
    case 3:
      return parseImageMessage(normalized)
    case 42:
@@ -490,8 +491,7 @@ function parseShareMessage(content: string): ParsedContent {
  }

  const articles = parseShareArticles(content)
-  const title =
-    articles[0]?.title || decodeXmlEntities(extractXmlValue(content, 'title')) || ''
+  const title = articles[0]?.title || decodeXmlEntities(extractXmlValue(content, 'title')) || ''
  const des =
    articles[0]?.description ||
    decodeXmlEntities(extractXmlValue(content, 'des') || extractXmlValue(content, 'desc')) ||
@@ -301,7 +301,7 @@ function listSourceMessages(
      /<appmsg\b|<refermsg\b|&lt;appmsg\b|&lt;refermsg\b/i.test(content)
        ? 49
        : msgType
-    if (!isPatMessage && [3, 42, 43, 47, 48, 49, 50, 10000, 10002].includes(inferredMsgType)) {
+    if (!isPatMessage && [3, 34, 42, 43, 47, 48, 49, 50, 10000, 10002].includes(inferredMsgType)) {
      try {
        const isQuotePayload = /<refermsg\b/i.test(content)
        const hasStickerPayload =
@@ -58,7 +58,8 @@ import type {
  VoiceModelDownloadResult,
  VoiceModelProgressEvent,
  VoiceModelStatus,
-  VoiceRecognitionResult
+  VoiceRecognitionResult,
+  VoiceTranscriptSnapshot
 } from '../shared/voice-recognition'
 import type {
  AiSearchCancelResult,
@@ -247,6 +248,9 @@ declare global {
      removeVoiceModel: () => Promise<VoiceModelStatus>
      openVoiceModelDirectory: () => Promise<{ success: boolean; error?: string }>
      recognizeVoice: (reference: VoiceMessageReference) => Promise<VoiceRecognitionResult>
+      getVoiceTranscriptSnapshot: (
+        reference: VoiceMessageReference
+      ) => Promise<VoiceTranscriptSnapshot>
      cancelVoiceRecognition: (reference: VoiceMessageReference) => Promise<{ success: boolean }>
      getVoiceBatchPreflight: (request: VoiceBatchRequest) => Promise<VoiceBatchPreflight>
      getVoiceBatchConversationSummaries: (
@@ -34,7 +34,8 @@ import type {
  VoiceModelDownloadResult,
  VoiceModelProgressEvent,
  VoiceModelStatus,
-  VoiceRecognitionResult
+  VoiceRecognitionResult,
+  VoiceTranscriptSnapshot
 } from '../shared/voice-recognition'
 import type {
  AiSearchCancelResult,
@@ -139,6 +140,10 @@ const api = {
    ipcRenderer.invoke('voice:openModelDirectory'),
  recognizeVoice: (reference: VoiceMessageReference): Promise<VoiceRecognitionResult> =>
    ipcRenderer.invoke('voice:recognize', reference),
+  getVoiceTranscriptSnapshot: (
+    reference: VoiceMessageReference
+  ): Promise<VoiceTranscriptSnapshot> =>
+    ipcRenderer.invoke('voice:getTranscriptSnapshot', reference),
  cancelVoiceRecognition: (reference: VoiceMessageReference): Promise<{ success: boolean }> =>
    ipcRenderer.invoke('voice:cancelRecognition', reference),
  getVoiceBatchPreflight: (request: VoiceBatchRequest): Promise<VoiceBatchPreflight> =>
@@ -387,6 +387,8 @@ export function useGroupReportGeneration({
            window.api.getVoiceModelStatus(),
            '检查语音模型'
          ) as Promise<VoiceModelStatus>,
+        getCachedTranscript: (reference) =>
+          withTimeout(window.api.getVoiceTranscriptSnapshot(reference), '读取语音缓存'),
        recognize: (reference) => withTimeout(window.api.recognizeVoice(reference), '语音转写'),
        onProgress: setVoiceTranscriptionProgress
      })
@@ -76,11 +76,15 @@ function friendlyImageNotice(warnings: string[]): string {
 export const isInternalIdentifier = (value: string): boolean =>
  /@chatroom$/i.test(value) || /^wxid_/i.test(value) || /^[a-z0-9_-]{18,}$/i.test(value)

+const isSystemMessage = (message: Message): boolean =>
+  message.from === 'system' || message.type === '系统消息' || message.contentData?.type === 'system'
+
 export const summarySender = (
  message: Message,
  contact: Contact | null,
  isGroup: boolean
 ): string => {
+  if (isSystemMessage(message)) return '微信系统消息'
  if (message.from === 'assistant') {
    const ownGroupNickname = message.name?.trim()
    if (isGroup && ownGroupNickname && !isInternalIdentifier(ownGroupNickname)) {
@@ -95,6 +99,11 @@ export const summarySender = (

 export const summaryContent = (message: Message): string => {
  const data = message.contentData
+  if (message.type === '语音' || data?.type === 'voice') {
+    return message.voiceTranscript?.trim()
+      ? `[语音${data?.type === 'voice' && data.duration ? ` ${data.duration}秒` : ''}] ${message.voiceTranscript.trim()}`
+      : `[语音${data?.type === 'voice' && data.duration ? ` ${data.duration}秒` : ''}]`
+  }
  if (!data) return message.content?.trim() || `[${message.type || '消息'}]`

  switch (data.type) {
@@ -102,10 +111,6 @@ export const summaryContent = (message: Message): string => {
      return '[图片]'
    case 'sticker':
      return '[表情]'
-    case 'voice':
-      return message.voiceTranscript?.trim()
-        ? `[语音${data.duration ? ` ${data.duration}秒` : ''}] ${message.voiceTranscript.trim()}`
-        : `[语音${data.duration ? ` ${data.duration}秒` : ''}]`
    case 'share':
      return data.articles?.length
        ? `[分享] ${data.articles
@@ -588,12 +593,14 @@ export const buildGroupReportFacts = async (
  for (const message of messages) {
    const sender = summarySender(message, contact, isGroup)
    const timestamp = parseTimestamp(message)
-    speakerCounts.set(sender, (speakerCounts.get(sender) || 0) + 1)
+    if (!isSystemMessage(message)) {
+      speakerCounts.set(sender, (speakerCounts.get(sender) || 0) + 1)
+      if (message.img && !avatars[sender]) avatars[sender] = message.img
+    }
    if (Number.isFinite(timestamp)) {
      const hour = new Date(timestamp).getHours()
      hourCounts.set(hour, (hourCounts.get(hour) || 0) + 1)
    }
-    if (message.img && !avatars[sender]) avatars[sender] = message.img
    if (message.contentData?.type === 'image') imageCount += 1
    if (message.contentData?.type === 'sticker') stickerCount += 1
    if (message.contentData?.type === 'voice') {
@@ -35,6 +35,7 @@ export const GROUP_REPORT_SYSTEM_PROMPT = `你是微信群聊日报编辑。请
 8. 所有候选条目尽量返回 sourceMessageIds，便于程序去重和追溯。
 9. 精简版面向 30 秒阅读，摘要必须短；完整版可以保留更多候选项。
 10. 只输出一个可被 JSON.parse 解析的 JSON 对象，不要输出 Markdown 代码块或其他文字。
+11. 发送者标记为“微信系统消息”的记录是平台通知，不是群成员；不得将其计入参与者、负责人、活跃成员或人物对话。

 JSON 结构必须为：
 {
@@ -2,7 +2,8 @@ import type { Message } from '../../../shared/types'
 import type {
  VoiceMessageReference,
  VoiceModelStatus,
-  VoiceRecognitionResult
+  VoiceRecognitionResult,
+  VoiceTranscriptSnapshot
 } from '../../../shared/voice-recognition'

 export interface VoiceTranscriptionProgress {
@@ -14,6 +15,7 @@ export interface VoiceTranscriptionProgress {

 interface VoiceTranscriptionDependencies {
  getModelStatus: () => Promise<VoiceModelStatus>
+  getCachedTranscript?: (reference: VoiceMessageReference) => Promise<VoiceTranscriptSnapshot>
  recognize: (reference: VoiceMessageReference) => Promise<VoiceRecognitionResult>
  onProgress: (progress: VoiceTranscriptionProgress) => void
 }
@@ -52,18 +54,12 @@ export async function transcribeVoiceMessages(
  }
  dependencies.onProgress({ ...progress })

-  const hasPendingVoice = voiceItems.some(
-    (item) => item.reference && !item.message.voiceTranscript?.trim()
-  )
-  if (hasPendingVoice) {
-    const modelStatus = await dependencies.getModelStatus()
-    if (modelStatus.state !== 'ready') {
-      throw new Error('请先在设置中准备离线语音识别模型，再生成包含语音转写的日报')
-    }
-  }
-
  const result = messages.map((message) => ({ ...message }))
+  const pendingItems: typeof voiceItems = []
  for (const item of voiceItems) {
+    if (!result[item.index].contentData) {
+      result[item.index].contentData = { type: 'voice' }
+    }
    const cachedTranscript = item.message.voiceTranscript?.trim()
    if (cachedTranscript) {
      result[item.index].voiceTranscript = cachedTranscript
@@ -72,19 +68,40 @@ export async function transcribeVoiceMessages(
      result[item.index].voiceTranscriptError = '语音标识不完整，无法定位本地语音'
      progress.failed += 1
    } else {
-      const recognition = await dependencies.recognize(item.reference)
-      const transcript = recognition.transcript?.trim()
-      if (recognition.success && transcript) {
-        result[item.index].voiceTranscript = transcript
+      const snapshot = await dependencies.getCachedTranscript?.(item.reference)
+      if (snapshot?.state === 'transcribed' && snapshot.transcript?.trim()) {
+        result[item.index].voiceTranscript = snapshot.transcript.trim()
        result[item.index].voiceTranscriptError = undefined
        progress.succeeded += 1
      } else {
-        result[item.index].voiceTranscriptError = recognition.error || '语音转写失败'
-        progress.failed += 1
+        pendingItems.push(item)
+        continue
      }
    }
    progress.processed += 1
    dependencies.onProgress({ ...progress })
  }
+
+  if (pendingItems.length) {
+    const modelStatus = await dependencies.getModelStatus()
+    if (modelStatus.state !== 'ready') {
+      throw new Error('请先在设置中准备离线语音识别模型，再生成包含语音转写的日报')
+    }
+  }
+
+  for (const item of pendingItems) {
+    const recognition = await dependencies.recognize(item.reference!)
+    const transcript = recognition.transcript?.trim()
+    if (recognition.success && transcript) {
+      result[item.index].voiceTranscript = transcript
+      result[item.index].voiceTranscriptError = undefined
+      progress.succeeded += 1
+    } else {
+      result[item.index].voiceTranscriptError = recognition.error || '语音转写失败'
+      progress.failed += 1
+    }
+    dependencies.onProgress({ ...progress, processed: progress.processed + 1 })
+    progress.processed += 1
+  }
  return result
 }
@@ -1,5 +1,9 @@
 export type ReportHeat = '高' | '中' | '低'
 export type ReportMode = 'compact' | 'full'
+
+export const selectHeroParticipantNames = (names: string[]): string[] =>
+  Array.from(new Set(names.map((name) => name.trim()).filter(Boolean))).slice(0, 4)
+
 export type ReportSectionKey =
  | 'hero'
  | 'topics'