feat(grok): 网关 Voice TTS/STT/Realtime 与分组音频定价

- 中继 xAI Voice HTTP（/tts /stt /custom-voices）与 Realtime WS（/realtime）
  仅 Grok 分组可用；账号选调度沿用 chat 能力，不依赖创作中心
- Voice URL 固定走 api.x.ai（CLI proxy base 自动回落）
- 分组级 audio_realtime / TTS / STT 单价：schema + migration 218 +
  admin API + GroupsView 表单与 i18n（无创作中心 UI）

frontend/src/i18n/locales/en/admin/overview.ts

@@ -994,6 +994,14 @@ export default {
         finalPricePreview: 'Final per-second price preview',
         notConfigured: 'Not configured'
       },
+      voicePricing: {
+        title: 'Grok Voice Pricing',
+        description: 'Optional per-group prices for Voice realtime / TTS / STT (USD). Leave empty to leave unpriced.',
+        audioRealtimePerMin: 'Realtime price per minute (USD)',
+        audioTtsPerMillionChars: 'TTS price per million chars (USD)',
+        audioSttPerHour: 'STT price per hour (USD)',
+        pricePlaceholder: 'optional'
+      },
       webSearchPricing: {
         title: 'Codex Web Search Pricing',
         pricePerCall: 'Price per search call (USD)',

frontend/src/i18n/locales/zh/admin/overview.ts

@@ -991,6 +991,14 @@ export default {
         finalPricePreview: '最终每秒价格预览',
         notConfigured: '未配置'
       },
+      voicePricing: {
+        title: 'Grok Voice 定价',
+        description: '分组级 Voice realtime / TTS / STT 单价（USD）。留空表示未配置。',
+        audioRealtimePerMin: 'Realtime 每分钟价格（USD）',
+        audioTtsPerMillionChars: 'TTS 每百万字符价格（USD）',
+        audioSttPerHour: 'STT 每小时价格（USD）',
+        pricePlaceholder: '可选'
+      },
       webSearchPricing: {
         title: 'Codex 网页搜索计费',
         pricePerCall: '搜索单次价格（USD/次）',

frontend/src/types/index.ts

@@ -572,6 +572,10 @@ export interface Group {
   video_model_prices?: VideoModelPrices
   // Codex 网页搜索单次价格（USD/次）；null 表示使用默认价 0.01
   web_search_price_per_call: number | null
+  // Grok Voice 显式定价（分组级）
+  audio_realtime_price_per_min: number | null
+  audio_tts_price_per_million_chars: number | null
+  audio_stt_price_per_hour: number | null
   // 高峰时段倍率配置
   peak_rate_enabled: boolean
   peak_start: string
@@ -772,6 +776,9 @@ export interface CreateGroupRequest {
   video_price_1080p?: number | null
   video_model_prices?: VideoModelPrices
   web_search_price_per_call?: number | null
+  audio_realtime_price_per_min?: number | null
+  audio_tts_price_per_million_chars?: number | null
+  audio_stt_price_per_hour?: number | null
   peak_rate_enabled?: boolean
   peak_start?: string
   peak_end?: string
@@ -828,6 +835,9 @@ export interface UpdateGroupRequest {
   video_price_1080p?: number | null
   video_model_prices?: VideoModelPrices
   web_search_price_per_call?: number | null
+  audio_realtime_price_per_min?: number | null
+  audio_tts_price_per_million_chars?: number | null
+  audio_stt_price_per_hour?: number | null
   peak_rate_enabled?: boolean
   peak_start?: string
   peak_end?: string

frontend/src/views/admin/GroupsView.vue

@@ -1483,6 +1483,57 @@
           </div>
         </div>
 
+
+        <!-- Grok Voice 显式定价（仅 grok 平台） -->
+        <div
+          v-if="createForm.platform === 'grok'"
+          class="border-t border-gray-200 dark:border-dark-400 pt-4 mt-4"
+        >
+          <h4 class="text-sm font-medium text-gray-700 dark:text-gray-300 mb-1">
+            {{ t("admin.groups.voicePricing.title") }}
+          </h4>
+          <p class="text-xs text-gray-500 dark:text-gray-400 mb-3">
+            {{ t("admin.groups.voicePricing.description") }}
+          </p>
+          <div class="grid grid-cols-1 gap-3 md:grid-cols-3">
+            <div>
+              <label class="input-label">{{ t("admin.groups.voicePricing.audioRealtimePerMin") }}</label>
+              <input
+                v-model.number="createForm.audio_realtime_price_per_min"
+                type="number"
+                step="0.000001"
+                min="0"
+                class="input"
+                :placeholder="t('admin.groups.voicePricing.pricePlaceholder')"
+                data-testid="create-audio-realtime-price"
+              />
+            </div>
+            <div>
+              <label class="input-label">{{ t("admin.groups.voicePricing.audioTtsPerMillionChars") }}</label>
+              <input
+                v-model.number="createForm.audio_tts_price_per_million_chars"
+                type="number"
+                step="0.000001"
+                min="0"
+                class="input"
+                :placeholder="t('admin.groups.voicePricing.pricePlaceholder')"
+                data-testid="create-audio-tts-price"
+              />
+            </div>
+            <div>
+              <label class="input-label">{{ t("admin.groups.voicePricing.audioSttPerHour") }}</label>
+              <input
+                v-model.number="createForm.audio_stt_price_per_hour"
+                type="number"
+                step="0.000001"
+                min="0"
+                class="input"
+                :placeholder="t('admin.groups.voicePricing.pricePlaceholder')"
+                data-testid="create-audio-stt-price"
+              />
+            </div>
+          </div>
+        </div>
         <!-- OpenAI Live 开关（仅 openai 平台） -->
         <div
           v-if="createForm.platform === 'openai'"
@@ -3123,6 +3174,57 @@
           </div>
         </div>
 
+
+        <!-- Grok Voice 显式定价（仅 grok 平台） -->
+        <div
+          v-if="editForm.platform === 'grok'"
+          class="border-t border-gray-200 dark:border-dark-400 pt-4 mt-4"
+        >
+          <h4 class="text-sm font-medium text-gray-700 dark:text-gray-300 mb-1">
+            {{ t("admin.groups.voicePricing.title") }}
+          </h4>
+          <p class="text-xs text-gray-500 dark:text-gray-400 mb-3">
+            {{ t("admin.groups.voicePricing.description") }}
+          </p>
+          <div class="grid grid-cols-1 gap-3 md:grid-cols-3">
+            <div>
+              <label class="input-label">{{ t("admin.groups.voicePricing.audioRealtimePerMin") }}</label>
+              <input
+                v-model.number="editForm.audio_realtime_price_per_min"
+                type="number"
+                step="0.000001"
+                min="0"
+                class="input"
+                :placeholder="t('admin.groups.voicePricing.pricePlaceholder')"
+                data-testid="edit-audio-realtime-price"
+              />
+            </div>
+            <div>
+              <label class="input-label">{{ t("admin.groups.voicePricing.audioTtsPerMillionChars") }}</label>
+              <input
+                v-model.number="editForm.audio_tts_price_per_million_chars"
+                type="number"
+                step="0.000001"
+                min="0"
+                class="input"
+                :placeholder="t('admin.groups.voicePricing.pricePlaceholder')"
+                data-testid="edit-audio-tts-price"
+              />
+            </div>
+            <div>
+              <label class="input-label">{{ t("admin.groups.voicePricing.audioSttPerHour") }}</label>
+              <input
+                v-model.number="editForm.audio_stt_price_per_hour"
+                type="number"
+                step="0.000001"
+                min="0"
+                class="input"
+                :placeholder="t('admin.groups.voicePricing.pricePlaceholder')"
+                data-testid="edit-audio-stt-price"
+              />
+            </div>
+          </div>
+        </div>
         <!-- OpenAI Live 开关（仅 openai 平台） -->
         <div
           v-if="editForm.platform === 'openai'"
@@ -4806,6 +4908,9 @@ const createForm = reactive({
   video_model_prices: createVideoModelPricesForm(),
   // Codex 网页搜索按次计费（仅 openai 平台使用）；null = 使用默认价 0.01
   web_search_price_per_call: null as number | null,
+  audio_realtime_price_per_min: null as number | null,
+  audio_tts_price_per_million_chars: null as number | null,
+  audio_stt_price_per_hour: null as number | null,
   // 高峰时段倍率配置
   peak_rate_enabled: false,
   peak_start: "",
@@ -5161,6 +5266,9 @@ const editForm = reactive({
   video_model_prices: createVideoModelPricesForm(),
   // Codex 网页搜索按次计费（仅 openai 平台使用）；null = 使用默认价 0.01
   web_search_price_per_call: null as number | null,
+  audio_realtime_price_per_min: null as number | null,
+  audio_tts_price_per_million_chars: null as number | null,
+  audio_stt_price_per_hour: null as number | null,
   // 高峰时段倍率配置
   peak_rate_enabled: false,
   peak_start: "",
@@ -5611,6 +5719,9 @@ const closeCreateModal = () => {
   createForm.video_price_1080p = null;
   createForm.video_model_prices = createVideoModelPricesForm();
   createForm.web_search_price_per_call = null;
+  createForm.audio_realtime_price_per_min = null;
+  createForm.audio_tts_price_per_million_chars = null;
+  createForm.audio_stt_price_per_hour = null;
   createForm.peak_rate_enabled = false;
   createForm.peak_start = "";
   createForm.peak_end = "";
@@ -5775,6 +5886,15 @@ const handleCreateGroup = async () => {
     requestData.video_price_480p = emptyToNull(requestData.video_price_480p);
     requestData.video_price_720p = emptyToNull(requestData.video_price_720p);
     requestData.video_price_1080p = emptyToNull(requestData.video_price_1080p);
+    requestData.audio_realtime_price_per_min = emptyToNull(
+      requestData.audio_realtime_price_per_min,
+    );
+    requestData.audio_tts_price_per_million_chars = emptyToNull(
+      requestData.audio_tts_price_per_million_chars,
+    );
+    requestData.audio_stt_price_per_hour = emptyToNull(
+      requestData.audio_stt_price_per_hour,
+    );
     requestData.web_search_price_per_call = emptyToNull(
       requestData.web_search_price_per_call,
     );
@@ -5835,6 +5955,9 @@ const handleEdit = async (group: AdminGroup) => {
     group.video_model_prices,
   );
   editForm.web_search_price_per_call = group.web_search_price_per_call ?? null;
+  editForm.audio_realtime_price_per_min = group.audio_realtime_price_per_min ?? null;
+  editForm.audio_tts_price_per_million_chars = group.audio_tts_price_per_million_chars ?? null;
+  editForm.audio_stt_price_per_hour = group.audio_stt_price_per_hour ?? null;
   editForm.peak_rate_enabled = group.peak_rate_enabled ?? false;
   editForm.peak_start = group.peak_start ?? "";
   editForm.peak_end = group.peak_end ?? "";
@@ -5916,6 +6039,9 @@ const closeEditModal = () => {
   editForm.video_price_1080p = null;
   editForm.video_model_prices = createVideoModelPricesForm();
   editForm.web_search_price_per_call = null;
+  editForm.audio_realtime_price_per_min = null;
+  editForm.audio_tts_price_per_million_chars = null;
+  editForm.audio_stt_price_per_hour = null;
   resetMessagesDispatchFormState(editForm);
   editForm.allow_live = false;
   resetModelsListState(editModelsListState);
@@ -6021,6 +6147,15 @@ const handleUpdateGroup = async () => {
     payload.video_price_480p = emptyPriceToClear(payload.video_price_480p);
     payload.video_price_720p = emptyPriceToClear(payload.video_price_720p);
     payload.video_price_1080p = emptyPriceToClear(payload.video_price_1080p);
+    payload.audio_realtime_price_per_min = emptyPriceToClear(
+      payload.audio_realtime_price_per_min,
+    );
+    payload.audio_tts_price_per_million_chars = emptyPriceToClear(
+      payload.audio_tts_price_per_million_chars,
+    );
+    payload.audio_stt_price_per_hour = emptyPriceToClear(
+      payload.audio_stt_price_per_hour,
+    );
     payload.web_search_price_per_call = emptyPriceToClear(
       payload.web_search_price_per_call,
     );