diff --git a/.env.example b/.env.example
index 1d854bc0e..cffee0eee 100644
--- a/.env.example
+++ b/.env.example
@@ -88,6 +88,9 @@ ASR_OPENAI_BASE_URL=
 ASR_QWEN_API_KEY=
 ASR_QWEN_BASE_URL=
 
+ASR_AZURE_API_KEY=
+ASR_AZURE_BASE_URL=https://{region}.api.cognitive.microsoft.com
+
 # --- PDF Processing -----------------------------------------------------------
 
 PDF_UNPDF_API_KEY=
diff --git a/components/settings/asr-settings.tsx b/components/settings/asr-settings.tsx
index 4a0cfb021..340a515de 100644
--- a/components/settings/asr-settings.tsx
+++ b/components/settings/asr-settings.tsx
@@ -33,6 +33,7 @@ export function ASRSettings({ selectedProviderId }: ASRSettingsProps) {
   const setASRProviderConfig = useSettingsStore((state) => state.setASRProviderConfig);
 
   const asrProvider = ASR_PROVIDERS[selectedProviderId] ?? ASR_PROVIDERS['openai-whisper'];
+  const availableModels = asrProvider.models ?? [];
   const isServerConfigured = !!asrProvidersConfig[selectedProviderId]?.isServerConfigured;
 
   const [showApiKey, setShowApiKey] = useState(false);
@@ -294,7 +295,7 @@ export function ASRSettings({ selectedProviderId }: ASRSettingsProps) {
       )}
 
       {/* Model Selection */}
-      {asrProvider.models.length > 0 && (
+      {availableModels.length > 0 && (
         <div className="space-y-2">
           <Label className="text-sm">{t('settings.ttsModel')}</Label>
           <Select
@@ -305,7 +306,7 @@ export function ASRSettings({ selectedProviderId }: ASRSettingsProps) {
               <SelectValue />
             </SelectTrigger>
             <SelectContent>
-              {asrProvider.models.map((model) => (
+              {availableModels.map((model) => (
                 <SelectItem key={model.id} value={model.id}>
                   {model.name}
                 </SelectItem>
diff --git a/components/settings/audio-settings.tsx b/components/settings/audio-settings.tsx
index d88590ac0..3983acb37 100644
--- a/components/settings/audio-settings.tsx
+++ b/components/settings/audio-settings.tsx
@@ -50,6 +50,7 @@ function getASRProviderName(providerId: ASRProviderId, t: (key: string) => strin
     'openai-whisper': t('settings.providerOpenAIWhisper'),
     'browser-native': t('settings.providerBrowserNative'),
     'qwen-asr': t('settings.providerQwenASR'),
+    'azure-asr': t('settings.providerAzureASR'),
   };
   return names[providerId];
 }
diff --git a/components/settings/index.tsx b/components/settings/index.tsx
index da9c11f56..1d5d721d2 100644
--- a/components/settings/index.tsx
+++ b/components/settings/index.tsx
@@ -134,6 +134,7 @@ function getASRProviderName(providerId: ASRProviderId, t: (key: string) => strin
     'openai-whisper': t('settings.providerOpenAIWhisper'),
     'browser-native': t('settings.providerBrowserNative'),
     'qwen-asr': t('settings.providerQwenASR'),
+    'azure-asr': t('settings.providerAzureASR'),
   };
   return names[providerId];
 }
diff --git a/lib/audio/asr-providers.ts b/lib/audio/asr-providers.ts
index 93365899a..d4302c9e0 100644
--- a/lib/audio/asr-providers.ts
+++ b/lib/audio/asr-providers.ts
@@ -184,6 +184,9 @@ export async function transcribeAudio(
     case 'qwen-asr':
       return await transcribeQwenASR(config, audioBuffer);
 
+    case 'azure-asr':
+      return await transcribeAzureASR(config, audioBuffer);
+
     default:
       throw new Error(`Unsupported ASR provider: ${config.providerId}`);
   }
@@ -326,6 +329,90 @@ async function transcribeQwenASR(
   return { text: transcribedText };
 }
 
+/**
+ * Azure STT implementation (Fast Transcription REST API)
+ * https://learn.microsoft.com/azure/ai-services/speech-service/fast-transcription-create
+ */
+async function transcribeAzureASR(
+  config: ASRModelConfig,
+  audioBuffer: Buffer | Blob,
+): Promise<ASRTranscriptionResult> {
+  const rawBaseUrl = config.baseUrl || ASR_PROVIDERS['azure-asr'].defaultBaseUrl!;
+
+  if (!rawBaseUrl || rawBaseUrl.includes('{region}')) {
+    throw new Error('Azure STT base URL must include a real region');
+  }
+
+  let endpoint = rawBaseUrl.replace(/\/+$/, '');
+  if (/\.stt\.speech\.microsoft\.com$/i.test(endpoint)) {
+    endpoint = endpoint.replace(/\.stt\.speech\.microsoft\.com$/i, '.api.cognitive.microsoft.com');
+  }
+  if (!/\/speechtotext\/transcriptions:transcribe/i.test(endpoint)) {
+    endpoint = `${endpoint}/speechtotext/transcriptions:transcribe`;
+  }
+  const url = new URL(endpoint);
+  if (!url.searchParams.get('api-version')) {
+    url.searchParams.set('api-version', '2025-10-15');
+  }
+
+  let audioBlob: Blob;
+  if (audioBuffer instanceof Blob) {
+    audioBlob = audioBuffer;
+  } else {
+    audioBlob = new Blob([audioBuffer as unknown as BlobPart], { type: 'audio/webm' });
+  }
+
+  const formData = new FormData();
+  formData.append('audio', audioBlob, 'recording.webm');
+
+  const localeMap: Record<string, string> = {
+    en: 'en-US',
+    zh: 'zh-CN',
+    ja: 'ja-JP',
+    ko: 'ko-KR',
+    de: 'de-DE',
+    fr: 'fr-FR',
+    es: 'es-ES',
+    it: 'it-IT',
+    pt: 'pt-BR',
+    ru: 'ru-RU',
+    ar: 'ar-SA',
+    hi: 'hi-IN',
+  };
+
+  if (config.language && config.language !== 'auto') {
+    const locale = localeMap[config.language] || config.language;
+    formData.append('definition', JSON.stringify({ locales: [locale] }));
+  }
+
+  const response = await fetch(url.toString(), {
+    method: 'POST',
+    headers: { 'Ocp-Apim-Subscription-Key': config.apiKey! },
+    body: formData,
+  });
+
+  if (!response.ok) {
+    const errorText = await response.text().catch(() => response.statusText);
+    throw new Error(`Azure STT error (${response.status}): ${errorText}`);
+  }
+
+  const data = (await response.json()) as {
+    combinedPhrases?: Array<{ text?: string }>;
+    phrases?: Array<{ text?: string }>;
+  };
+
+  const combinedText = data.combinedPhrases
+    ?.map((p) => p.text || '')
+    .filter(Boolean)
+    .join(' ');
+  const phraseText = data.phrases
+    ?.map((p) => p.text || '')
+    .filter(Boolean)
+    .join(' ');
+
+  return { text: combinedText || phraseText || '' };
+}
+
 /**
  * Get current ASR configuration from settings store
  * Note: This function should only be called in browser context
diff --git a/lib/audio/constants.ts b/lib/audio/constants.ts
index 423f5b82c..2f580ab29 100644
--- a/lib/audio/constants.ts
+++ b/lib/audio/constants.ts
@@ -1031,6 +1031,32 @@ export const ASR_PROVIDERS: Record<ASRProviderId, ASRProviderConfig> = {
     supportedFormats: ['mp3', 'wav', 'webm', 'm4a', 'flac'],
   },
 
+  'azure-asr': {
+    id: 'azure-asr',
+    name: 'Azure STT',
+    requiresApiKey: true,
+    defaultBaseUrl: 'https://{region}.api.cognitive.microsoft.com',
+    icon: '/logos/azure.svg',
+    models: [],
+    defaultModelId: '',
+    supportedLanguages: [
+      'auto',
+      'en',
+      'zh',
+      'ja',
+      'ko',
+      'de',
+      'fr',
+      'es',
+      'it',
+      'pt',
+      'ru',
+      'ar',
+      'hi',
+    ],
+    supportedFormats: ['wav', 'ogg', 'webm', 'mp3', 'flac', 'm4a'],
+  },
+
   'browser-native': {
     id: 'browser-native',
     name: '浏览器原生 ASR (Web Speech API)',
diff --git a/lib/audio/types.ts b/lib/audio/types.ts
index 0c3c91792..d808a021c 100644
--- a/lib/audio/types.ts
+++ b/lib/audio/types.ts
@@ -16,6 +16,7 @@
  * - OpenAI Whisper (https://platform.openai.com/docs/guides/speech-to-text)
  * - Browser Native (Web Speech API, client-side only)
  * - Qwen ASR (DashScope API)
+ * - Azure STT (https://learn.microsoft.com/azure/ai-services/speech-service/fast-transcription-create)
  *
  * Future Provider Support (extensible):
  * - ElevenLabs TTS/ASR (https://elevenlabs.io/docs)
@@ -152,7 +153,7 @@ export interface TTSModelConfig {
  * Add new ASR providers here as union members.
  * Keep in sync with ASR_PROVIDERS registry in constants.ts
  */
-export type ASRProviderId = 'openai-whisper' | 'browser-native' | 'qwen-asr';
+export type ASRProviderId = 'openai-whisper' | 'browser-native' | 'qwen-asr' | 'azure-asr';
 // Add new ASR providers below (uncomment and modify):
 // | 'elevenlabs-asr'
 // | 'assemblyai-asr'
diff --git a/lib/i18n/locales/en-US.json b/lib/i18n/locales/en-US.json
index 997857983..1f9a1b90c 100644
--- a/lib/i18n/locales/en-US.json
+++ b/lib/i18n/locales/en-US.json
@@ -539,6 +539,7 @@
     "providerOpenAIWhisper": "OpenAI ASR (gpt-4o-mini-transcribe)",
     "providerBrowserNative": "Browser Native ASR",
     "providerQwenASR": "Qwen ASR (Alibaba Cloud Bailian)",
+    "providerAzureASR": "Azure STT",
     "providerUnpdf": "unpdf (Built-in)",
     "providerMinerU": "MinerU",
     "browserNativeTTSNote": "Browser Native TTS requires no configuration and is completely free, using system built-in voices",
diff --git a/lib/i18n/locales/ja-JP.json b/lib/i18n/locales/ja-JP.json
index 6d82ebf07..c789cb617 100644
--- a/lib/i18n/locales/ja-JP.json
+++ b/lib/i18n/locales/ja-JP.json
@@ -539,6 +539,7 @@
     "providerOpenAIWhisper": "OpenAI ASR (gpt-4o-mini-transcribe)",
     "providerBrowserNative": "ブラウザネイティブASR",
     "providerQwenASR": "Qwen ASR（Alibaba Cloud百錬）",
+    "providerAzureASR": "Azure STT",
     "providerUnpdf": "unpdf（組み込み）",
     "providerMinerU": "MinerU",
     "browserNativeTTSNote": "ブラウザネイティブTTSは設定不要で完全無料です。システム内蔵のボイスを使用します",
diff --git a/lib/i18n/locales/ru-RU.json b/lib/i18n/locales/ru-RU.json
index 9afa76e68..32a1230f6 100644
--- a/lib/i18n/locales/ru-RU.json
+++ b/lib/i18n/locales/ru-RU.json
@@ -539,6 +539,7 @@
     "providerOpenAIWhisper": "OpenAI ASR (gpt-4o-mini-transcribe)",
     "providerBrowserNative": "Встроенный ASR браузера",
     "providerQwenASR": "Qwen ASR (Alibaba Cloud Bailian)",
+    "providerAzureASR": "Azure STT",
     "providerUnpdf": "unpdf (встроенный)",
     "providerMinerU": "MinerU",
     "browserNativeTTSNote": "Встроенный TTS браузера не требует настройки и полностью бесплатен, использует системные голоса",
diff --git a/lib/i18n/locales/zh-CN.json b/lib/i18n/locales/zh-CN.json
index b2f7a2a79..d3f639a37 100644
--- a/lib/i18n/locales/zh-CN.json
+++ b/lib/i18n/locales/zh-CN.json
@@ -539,6 +539,7 @@
     "providerOpenAIWhisper": "OpenAI ASR (gpt-4o-mini-transcribe)",
     "providerBrowserNative": "浏览器原生 ASR",
     "providerQwenASR": "Qwen ASR（阿里云百炼）",
+    "providerAzureASR": "Azure STT",
     "providerUnpdf": "unpdf（内置）",
     "providerMinerU": "MinerU",
     "browserNativeTTSNote": "浏览器原生 TTS 无需配置，完全免费，使用系统内置语音",
diff --git a/lib/server/provider-config.ts b/lib/server/provider-config.ts
index 259208fde..fb4fce446 100644
--- a/lib/server/provider-config.ts
+++ b/lib/server/provider-config.ts
@@ -65,6 +65,7 @@ const TTS_ENV_MAP: Record<string, string> = {
 const ASR_ENV_MAP: Record<string, string> = {
   ASR_OPENAI: 'openai-whisper',
   ASR_QWEN: 'qwen-asr',
+  ASR_AZURE: 'azure-asr',
 };
 
 const PDF_ENV_MAP: Record<string, string> = {
diff --git a/lib/store/settings.ts b/lib/store/settings.ts
index 4b088bbc6..a68205e14 100644
--- a/lib/store/settings.ts
+++ b/lib/store/settings.ts
@@ -299,6 +299,7 @@ const getDefaultAudioConfig = () => ({
     'openai-whisper': { apiKey: '', baseUrl: '', enabled: true },
     'browser-native': { apiKey: '', baseUrl: '', enabled: true },
     'qwen-asr': { apiKey: '', baseUrl: '', enabled: false },
+    'azure-asr': { apiKey: '', baseUrl: '', enabled: false },
   } as Record<ASRProviderId, { apiKey: string; baseUrl: string; enabled: boolean }>,
 });