diff --git a/README.md b/README.md
index f50ec73..7a6663d 100644
--- a/README.md
+++ b/README.md
@@ -146,6 +146,17 @@ xattr -rd com.apple.quarantine /path/to/Polyglot.app
 + 选择“F0”价格层，并单击“创建”
 + 创建完成后，转到新服务的“概述”页面，即可查看密钥和终结点
 
+</details>
+<details>
+<summary>3. aws语音服务如何申请？</summary>
+
+要申请 AWS 语音服务的 Identity Pool ID ，可以按照以下步骤进行：
+
++ 登录 AWS 门户 （https://console.aws.amazon.com/）
++ 转到“Cognito 认证服务”页面并单击“Create Identity Pool”按钮
++ 在“IAM Role”窗口中，创建Role, 添加Polly, Transcribe 权限即可
++ 创建完成后，转到“概述”页面，即可Identity Pool ID
+
 </details>
 
 <!-- ## 捐赠
diff --git a/package.json b/package.json
index 5763612..1a9a433 100644
--- a/package.json
+++ b/package.json
@@ -34,13 +34,19 @@
     "release:version": "npx standard-version && git push origin --follow-tags"
   },
   "dependencies": {
+    "@aws-sdk/client-cognito-identity": "^3.363.0",
+    "@aws-sdk/client-polly": "^3.363.0",
+    "@aws-sdk/client-transcribe-streaming": "^3.363.0",
+    "@aws-sdk/credential-provider-cognito-identity": "^3.363.0",
     "@iconify-json/svg-spinners": "^1.1.1",
     "@vueuse/core": "^9.13.0",
     "api2d": "^0.1.18",
+    "aws-sdk": "^2.1409.0",
     "dexie": "^3.2.3",
     "electron-updater": "^5.3.0",
     "element-plus": "^2.3.3",
     "eventsource-parser": "^0.1.0",
+    "microphone-stream": "^6.0.1",
     "microsoft-cognitiveservices-speech-sdk": "^1.26.0",
     "pinia": "^2.0.33",
     "pinia-plugin-persistedstate": "^3.1.0",
diff --git a/src/config.ts b/src/config.ts
index 0a9483d..6fd9aa6 100644
--- a/src/config.ts
+++ b/src/config.ts
@@ -190,6 +190,16 @@ export const supportLanguageMap = {
   'zh-TW': '中文(台湾普通话)',
 } as Record<string, string>
 
+export const awsRegions = [
+  'us-east-1',
+  'us-east-2',
+  'us-west-1',
+  'us-west-2',
+  'ap-east-1',
+  'ap-southeast-1',
+  'eu-central-1',
+]
+
 export const azureRegions = [
   'australiaeast',
   'australiasoutheast',
diff --git a/src/constant.ts b/src/constant.ts
index 952630a..d29e8f7 100644
--- a/src/constant.ts
+++ b/src/constant.ts
@@ -2,6 +2,8 @@ export const OPEN_KEY = 'openKey'
 export const OPEN_PROXY = 'openProxy'
 export const AZURE_REGION = 'azureRegion'
 export const AZURE_KEY = 'azureKey'
+export const AWS_REGION = 'awsRegion'
+export const AWS_COGNITO_IDENTITY_POOL_ID = 'awsCognitoIdentityPoolId'
 export const AZURE_TRANSLATE_KEY = 'azureTranslateKey'
 export const VOICE_API_NAME = 'voiceApiName'
 export const IS_ALWAYS_RECOGNITION = 'isAlwaysRecognition'
diff --git a/src/hooks/useGlobalSetting.ts b/src/hooks/useGlobalSetting.ts
index e78b687..375772a 100644
--- a/src/hooks/useGlobalSetting.ts
+++ b/src/hooks/useGlobalSetting.ts
@@ -1,4 +1,4 @@
-import { AUTO_PLAY, AZURE_KEY, AZURE_REGION, AZURE_TRANSLATE_KEY, CHAT_API_NAME, CHAT_REMEMBER_COUNT, IS_ALWAYS_RECOGNITION, OPEN_KEY, OPEN_MAX_TOKEN, OPEN_MODEL, OPEN_PROXY, SELF_AVATAR_URL, TTS_PASSWORD, VOICE_API_NAME } from '@/constant'
+import { AUTO_PLAY,AWS_COGNITO_IDENTITY_POOL_ID, AWS_KEY,AWS_REGION,AWS_SECRET_KEY,AZURE_KEY, AZURE_REGION, AZURE_TRANSLATE_KEY, CHAT_API_NAME, CHAT_REMEMBER_COUNT, IS_ALWAYS_RECOGNITION, OPEN_KEY, OPEN_MAX_TOKEN, OPEN_MODEL, OPEN_PROXY, SELF_AVATAR_URL, TTS_PASSWORD, VOICE_API_NAME } from '@/constant'
 
 import { getAvatarUrl } from '@/utils'
 
@@ -7,6 +7,9 @@ export const useGlobalSetting = () => {
   const openProxy = useLocalStorage(OPEN_PROXY, '')
   const azureRegion = useLocalStorage(AZURE_REGION, 'eastasia')
   const azureKey = useLocalStorage(AZURE_KEY, '')
+  const awsRegion = useLocalStorage(AWS_REGION, 'us-east-1')
+  const awsCognitoIdentityId = useLocalStorage(AWS_COGNITO_IDENTITY_POOL_ID, '')
+
   const openModel = useLocalStorage(OPEN_MODEL, 'gpt-3.5-turbo')
   const selfAvatar = useLocalStorage(SELF_AVATAR_URL, getAvatarUrl('self.png'))
   const chatApiName = useLocalStorage(CHAT_API_NAME, 'openAI')
@@ -22,6 +25,8 @@ export const useGlobalSetting = () => {
     openKey,
     openProxy,
     openModel,
+    awsRegion,
+    awsCognitoIdentityId,
     azureRegion,
     azureKey,
     selfAvatar,
diff --git a/src/hooks/useSpeechService.ts b/src/hooks/useSpeechService.ts
index 1af5a26..f1a36c5 100644
--- a/src/hooks/useSpeechService.ts
+++ b/src/hooks/useSpeechService.ts
@@ -8,6 +8,16 @@ import {
   SpeechSynthesizer,
 } from 'microsoft-cognitiveservices-speech-sdk'
 
+import MicrophoneStream from 'microphone-stream';
+import { CognitoIdentityClient } from "@aws-sdk/client-cognito-identity";
+import {fromCognitoIdentityPool} from "@aws-sdk/credential-provider-cognito-identity";
+import { Polly,SynthesizeSpeechInput,DescribeVoicesCommand } from "@aws-sdk/client-polly";
+import {
+  TranscribeStreamingClient,
+  StartStreamTranscriptionCommand,
+} from '@aws-sdk/client-transcribe-streaming';
+
+
 const defaultAzureRegion = import.meta.env.VITE_REGION
 const defaultAzureKey = import.meta.env.VITE_SCRIPTION_KEY
 const accessPassword = import.meta.env.VITE_TTS_ACCESS_PASSWORD
@@ -17,8 +27,13 @@ interface Config {
   isFetchAllVoice?: boolean
 }
 export const useSpeechService = ({ langs = <const>['fr-FR', 'ja-JP', 'en-US', 'zh-CN', 'zh-HK', 'ko-KR', 'de-DE'], isFetchAllVoice = true }: Config = {}) => {
-  const { azureKey, azureRegion, ttsPassword } = useGlobalSetting()
+  const { azureKey, azureRegion, ttsPassword,voiceApiName } = useGlobalSetting()
+  const { awsCognitoIdentityId, awsRegion, } = useGlobalSetting()
 
+
+  if(voiceApiName.value==="AWS"){
+    isFetchAllVoice=false;
+  }
   const resultAzureKey = computed(() => {
     if (!azureKey.value) {
       if (accessPassword !== ttsPassword.value)
@@ -58,6 +73,7 @@ export const useSpeechService = ({ langs = <const>['fr-FR', 'ja-JP', 'en-US', 'z
   const audioBlob = ref<Blob>(new Blob())
 
   const allVoices = ref<VoiceInfo[]>([])
+  const allAWSVoices = ref<any[]>([])
 
   const recognizer = ref<SpeechRecognizer>(new SpeechRecognizer(speechConfig.value))
   const synthesizer = ref<SpeechSynthesizer>(new SpeechSynthesizer(speechConfig.value))
@@ -74,8 +90,28 @@ export const useSpeechService = ({ langs = <const>['fr-FR', 'ja-JP', 'en-US', 'z
     immediate: true,
   })
 
-  // 语音识别
 
+  // AWS polly and transcribe SDK 初始化
+  const audioAWS = new Audio();
+  let micStream: MicrophoneStream | undefined = undefined
+  const polly = new Polly({
+    region: awsRegion.value ?? "us-east-1",
+    credentials: fromCognitoIdentityPool({
+      client: new CognitoIdentityClient({ region: awsRegion.value ?? "us-east-1" }),
+      identityPoolId: awsCognitoIdentityId.value
+    }),
+  });
+
+  const transcribe = new TranscribeStreamingClient({
+    region: awsRegion.value ?? "us-east-1",
+    credentials: fromCognitoIdentityPool({
+      client: new CognitoIdentityClient({ region: awsRegion.value ?? "us-east-1" }),
+      identityPoolId: awsCognitoIdentityId.value
+    }),
+  });
+
+
+  // AZure 语音识别
   const audioRecorder = async () => {
     // 暂时通过 mediaRecorder 方式实现录音保存，后续可能会改为直接通过 SpeechRecognizer 实现保存
 
@@ -250,16 +286,41 @@ export const useSpeechService = ({ langs = <const>['fr-FR', 'ja-JP', 'en-US', 'z
       catch (error) {
         allVoices.value = []
       }
+      const res = await synthesizer.value.getVoicesAsync()
+      if (res.errorDetails) {
+        console.error(`获取语音列表失败：${res.errorDetails}, 请检查语音配置`)
+        return []
+      }
+      return res.voices
+    }else{
+      return []
     }
 
-    const res = await synthesizer.value.getVoicesAsync()
-    if (res.errorDetails) {
-      console.error(`获取语音列表失败：${res.errorDetails}, 请检查语音配置`)
-      return []
+   
+  }
+
+  // 获取AWS 语音列表
+  async function getAWSVoices() {
+    const params = {
+      LanguageCode: "en-US"
+    };
+  
+    try {
+      const data = await polly.describeVoices(params)
+      if(data.Voices){
+        allAWSVoices.value=data.Voices.map((item)=>{
+          return {"id":item.Id,"gender":item.Gender}
+        })
+      }  
+      return data.Voices??[];
+    } catch (error) {
+      console.error("Error retrieving AWS voices:", error);
+      return [];
     }
-    return res.voices
+
   }
 
+
   function applySynthesizerConfiguration() {
     // 通过playback结束事件来判断播放结束
     player.value = new SpeakerAudioDestination()
@@ -279,6 +340,100 @@ export const useSpeechService = ({ langs = <const>['fr-FR', 'ja-JP', 'en-US', 'z
     synthesizer.value = new SpeechSynthesizer(speechConfig.value, speakConfig)
   }
 
+  /* AWS Vocie service */
+  const startAWSRecognizeSpeech = async (cb?: (text: string) => void) => {
+
+    micStream = new MicrophoneStream();
+    // // this part should be put into an async function
+
+    micStream.setStream(
+      await window.navigator.mediaDevices.getUserMedia({
+        video: false,
+        audio: true,
+      })
+    );
+
+
+    //构造audioSream
+    isRecognizing.value = true
+    const MAX_AUDIO_CHUNK_SIZE = 48000
+
+    const audioStream = async function* () {
+      for await (const chunk of micStream as unknown as Iterable<Buffer>) {
+        if (chunk.length <= MAX_AUDIO_CHUNK_SIZE) {
+          yield {
+            AudioEvent: {
+              AudioChunk: pcmEncodeChunk(chunk),
+            },
+          }
+        }
+      }
+    };
+
+    //PCM 编码
+    const pcmEncodeChunk = (chunk: any) => {
+      const input = MicrophoneStream.toRaw(chunk);
+      var offset = 0;
+      var buffer = new ArrayBuffer(input.length * 2);
+      var view = new DataView(buffer);
+      for (var i = 0; i < input.length; i++, offset += 2) {
+        var s = Math.max(-1, Math.min(1, input[i]));
+        view.setInt16(offset, s < 0 ? s * 0x8000 : s * 0x7fff, true);
+      }
+      return Buffer.from(buffer);
+    };
+    //Transcribe stream command 初始化
+    const command = new StartStreamTranscriptionCommand({
+      LanguageCode: language.value,
+      MediaEncoding: "pcm",
+      MediaSampleRateHertz: 44100,
+      AudioStream: audioStream(),
+    });
+
+    const response = await transcribe.send(command);
+    let resultText = ""
+    if (response.TranscriptResultStream) {
+      for await (const event of response.TranscriptResultStream) {
+        if (event.TranscriptEvent) {
+          const results = event.TranscriptEvent?.Transcript?.Results;
+          results?.map((result: any) => {
+            (result.Alternatives || []).map((alternative: any) => {
+              const transcript = alternative.Items.map((item: any) => item.Content).join(" ");
+              resultText = transcript;
+              cb && cb(transcript)
+            });
+          });
+        }
+      }
+      isRecognizing.value = false
+    }
+    return resultText
+
+  }
+
+  const stopAWSRecognizeSpeech = () => {
+    micStream?.stop()
+  }
+
+
+   //语音合成
+   const awsTextToSpeak = async (text: string, voice?: string) => {
+    const params: SynthesizeSpeechInput = {
+      Text: text,
+      OutputFormat: 'mp3',
+      VoiceId: 'Joanna', // Replace with the desired voice ID (e.g., Joanna, Matthew, etc.)
+    };
+
+    const response = await polly.synthesizeSpeech(params);
+
+    if (response.AudioStream) {
+      const buffer = await response.AudioStream.transformToByteArray();
+      audioAWS.src = URL.createObjectURL(new Blob([buffer], { type: 'audio/mpeg' }));
+      audioAWS.play();
+    }
+  }
+  
+
   return {
     languages,
     language,
@@ -289,16 +444,23 @@ export const useSpeechService = ({ langs = <const>['fr-FR', 'ja-JP', 'en-US', 'z
     isRecognizReadying,
     startRecognizeSpeech,
     stopRecognizeSpeech,
+    startAWSRecognizeSpeech,
+    stopAWSRecognizeSpeech,
     recognizeSpeech,
     textToSpeak,
+    awsTextToSpeak,
     ssmlToSpeak,
     stopTextToSpeak,
     getVoices,
+    getAWSVoices,
     allVoices,
+    allAWSVoices,
     isSynthesizing,
     rate,
     style,
     audioBlob,
     player,
+    audioAWS,
+    
   }
 }
diff --git a/src/pages/Home/components/Content.vue b/src/pages/Home/components/Content.vue
index 8f05d5f..da6b48b 100644
--- a/src/pages/Home/components/Content.vue
+++ b/src/pages/Home/components/Content.vue
@@ -14,7 +14,7 @@ interface Translates {
 // hooks
 const store = useConversationStore()
 const { el, scrollToBottom } = useScroll()
-const { selfAvatar, openKey, chatRememberCount, autoPlay } = useGlobalSetting()
+const { selfAvatar, voiceApiName, openKey, chatRememberCount, autoPlay } = useGlobalSetting()
 
 const {
   language,
@@ -26,12 +26,19 @@ const {
   startRecognizeSpeech,
   isRecognizReadying,
   stopRecognizeSpeech,
+  startAWSRecognizeSpeech,
+  stopAWSRecognizeSpeech,
   ssmlToSpeak,
+  awsTextToSpeak,
   isSynthesizing,
   audioBlob,
   player,
+  audioAWS,
 } = useSpeechService({ langs: store.allLanguage as any, isFetchAllVoice: false })
 
+
+
+
 // states
 const message = ref('') // input message
 const text = ref('') // current select message
@@ -127,7 +134,13 @@ const fetchResponse = async (prompt: ChatMessage[] | string, type: FetchType = F
 }
 
 async function onSubmit(fromRecognize = false) {
-  if (!verifyOpenKey(openKey.value)) return alert('请输入正确的API-KEY')
+  //增加OPENAPI Proxy 验证模式:  OPENAPI_KEY:PROXY_KEY 
+  if (openKey.value.indexOf(":") > -1) {
+    if (!verifyOpenKey(openKey.value.split(":")[0])) return alert('请输入正确的API-KEY')
+  } else {
+    if (!verifyOpenKey(openKey.value)) return alert('请输入正确的API-KEY')
+  }
+
   if (!message.value) return
 
   store.changeConversations([
@@ -210,24 +223,54 @@ function speakByAI(content: string, index: number) {
   }
   speakIndex.value = index
   text.value = content
-  ssmlToSpeak(content)
+  
+  if (voiceApiName.value === "AWS") {
+    awsTextToSpeak(content)
+  } else {
+    ssmlToSpeak(content)
+  }
+
 }
 
+
+
 const recognize = async () => {
   try {
     console.log('isRecognizing', isRecognizing.value)
-    if (isRecognizing.value) {
-      await stopRecognizeSpeech()
-      onSubmit(true)
-      console.log('submit', message.value)
-      return
+    if (voiceApiName.value === "AWS") {
+      if (isRecognizing.value){
+        //延迟关闭AWS Transcribe Stream Client ,避免语音未完整上传
+        setTimeout(()=>{
+            stopAWSRecognizeSpeech()
+        },2000)
+        isRecognizing.value=false
+      }else{
+        audioAWS.pause()
+        message.value = ''
+      
+   
+      stopAllSpeaker() // 开启语音识别时停止所有语音播放
+      startAWSRecognizeSpeech((textSlice:string) => {
+        message.value =textSlice || ''
+      });
+      
+    }
+     
+    } else {
+      if (isRecognizing.value) {
+        await stopRecognizeSpeech()
+        onSubmit(true)
+        console.log('submit', message.value)
+        return
+      }
+      message.value = ''
+
+      stopAllSpeaker() // 开启语音识别时停止所有语音播放
+      startRecognizeSpeech((textSlice) => {
+        message.value += textSlice || ''
+      })
     }
-    message.value = ''
 
-    stopAllSpeaker() // 开启语音识别时停止所有语音播放
-    startRecognizeSpeech((textSlice) => {
-      message.value += textSlice || ''
-    })
   }
   catch (error) {
     alert(error)
@@ -277,12 +320,8 @@ async function grammarAnalysis(text: string, i: number) {
   <div flex flex-col p-2 rounded-md bg-white dark="bg-#1e1e1e" shadow-sm>
     <div ref="el" class="hide-scrollbar flex-1 overflow-auto">
       <template v-if="chatMessages.length">
-        <div
-          v-for="item, i in chatMessages"
-          :key="i"
-          center-y
-          :class="item.role === 'user' ? 'flex-row-reverse w-[75%] ml-[25%]' : 'w-[75%]'"
-        >
+        <div v-for="item, i in chatMessages" :key="i" center-y
+          :class="item.role === 'user' ? 'flex-row-reverse w-[75%] ml-[25%]' : 'w-[75%]'">
           <div class="w-10 h-10">
             <img w-full h-full object-fill rounded-full :src="item.role === 'user' ? selfAvatar : currentAvatar" alt="">
           </div>
@@ -290,11 +329,11 @@ async function grammarAnalysis(text: string, i: number) {
             <p p-2 my-2 chat-box>
               {{ item.content }}
             </p>
-            <p v-show=" translates[item.content + i]?.isShow " p-2 my-2 chat-box>
+            <p v-show="translates[item.content + i]?.isShow" p-2 my-2 chat-box>
               {{ translates[item.content + i]?.result }}
             </p>
 
-            <p v-show="item.role === 'user' && analysisResults[item.content + i]?.isShow " p-2 my-2 chat-box>
+            <p v-show="item.role === 'user' && analysisResults[item.content + i]?.isShow" p-2 my-2 chat-box>
               {{ analysisResults[item.content + i]?.result }}
             </p>
 
@@ -313,7 +352,8 @@ async function grammarAnalysis(text: string, i: number) {
                   <i icon-btn rotate-90 i-ic:sharp-wifi />
                 </span>
               </template>
-              <span v-if="!isTranslating || translateIndex !== i" ml-1 class="chat-btn" @click="translate(item.content, i)">
+              <span v-if="!isTranslating || translateIndex !== i" ml-1 class="chat-btn"
+                @click="translate(item.content, i)">
                 <i icon-btn i-carbon:ibm-watson-language-translator />
               </span>
               <span v-else ml-1 class="chat-btn">
@@ -338,7 +378,8 @@ async function grammarAnalysis(text: string, i: number) {
                   </span>
                 </template>
               </template>
-              <span v-if="!analysisLoading || analysisIndex !== i" class="chat-btn" @click="grammarAnalysis(item.content, i)">
+              <span v-if="!analysisLoading || analysisIndex !== i" class="chat-btn"
+                @click="grammarAnalysis(item.content, i)">
                 <i icon-btn i-ic:outline-lightbulb />
               </span>
               <span v-else class="chat-btn">
@@ -356,11 +397,8 @@ async function grammarAnalysis(text: string, i: number) {
     </div>
 
     <div class="flex h-10 w-[-webkit-fill-available] mt-1">
-      <Button
-        mr-1
-        :disabled="isRecognizReadying || store.loading"
-        @click="recognize()"
-      >
+      <!-- <audio id="player" controls></audio> -->
+      <Button mr-1 :disabled="isRecognizReadying || store.loading" @click="recognize()">
         <i v-if="isRecognizReadying" i-eos-icons:bubble-loading />
         <i v-else i-carbon:microphone />
       </Button>
@@ -373,30 +411,17 @@ async function grammarAnalysis(text: string, i: number) {
         录音设备准备中
         <i icon-btn i-eos-icons:three-dots-loading />
       </div>
-      <input
-        v-else-if="!store.loading"
-        v-model="message"
-        type="text"
-        placeholder="Type your message here..."
-        input-box
-        p-3 flex-1
-        @blur="store.changeMainActive(true)" @focus="store.changeMainActive(false)" @keypress.enter="() => onSubmit()"
-      >
+      <input v-else-if="!store.loading" v-model="message" type="text" placeholder="Type your message here..." input-box
+        p-3 flex-1 @blur="store.changeMainActive(true)" @focus="store.changeMainActive(false)"
+        @keypress.enter="() => onSubmit()">
       <div v-else class="loading-btn">
         AI Is Thinking
         <i icon-btn i-eos-icons:three-dots-loading />
       </div>
-      <Button
-        :disabled="store.loading"
-        mx-1
-        @click="onSubmit"
-      >
+      <Button :disabled="store.loading" mx-1 @click="onSubmit">
         <i i-carbon:send-alt />
       </Button>
-      <Button
-        :disabled="store.loading"
-        @click="store.cleanCurrentConversations()"
-      >
+      <Button :disabled="store.loading" @click="store.cleanCurrentConversations()">
         <i i-carbon:trash-can />
       </Button>
     </div>
diff --git a/src/pages/Setting/components/TTSSetting.vue b/src/pages/Setting/components/TTSSetting.vue
index fe776ac..38a3bff 100644
--- a/src/pages/Setting/components/TTSSetting.vue
+++ b/src/pages/Setting/components/TTSSetting.vue
@@ -1,6 +1,6 @@
 <script setup lang="ts">
-import { azureRegions } from '@/config'
-const { azureRegion, azureKey, isAlwaysRecognition, ttsPassword, autoPlay, voiceApiName } = useGlobalSetting()
+import { azureRegions, awsRegions } from '@/config'
+const { awsRegion,awsCognitoIdentityId,awsSecretKey, azureRegion, azureKey, isAlwaysRecognition, ttsPassword, autoPlay, voiceApiName } = useGlobalSetting()
 </script>
 
 <template>
@@ -12,7 +12,7 @@ const { azureRegion, azureKey, isAlwaysRecognition, ttsPassword, autoPlay, voice
           <el-tooltip
             class="box-item"
             effect="dark"
-            content="目前仅支持Azure服务"
+            content="目前仅支持Azure, AWS服务"
             placement="bottom"
           >
             <i icon-btn i-carbon:information-square />
@@ -25,9 +25,36 @@ const { azureRegion, azureKey, isAlwaysRecognition, ttsPassword, autoPlay, voice
           <option value="Azure">
             Azure
           </option>
+          <option value="AWS">
+            AWS
+          </option>
         </select>
       </div>
-      <div class="section-item">
+      <div class="section-item" v-show="voiceApiName=='AWS'">
+        <label my-1 block for="">AWS 区域</label>
+        <select
+          v-model="awsRegion"
+        >
+          <option v-for="item in awsRegions" :key="item" :value="item">
+            {{ item }}
+          </option>
+        </select>
+      </div>
+      <div class="section-item" v-show="voiceApiName=='AWS'">
+        <div center-y>
+          <label mr-1 my-1 block for="">AWS Cognito Identity Pool ID</label>
+          <el-tooltip
+            effect="dark"
+            content="使用自己的AWS Cognito Identity Pool ID"
+            placement="bottom"
+          >
+            <i icon-btn i-carbon:information-square />
+          </el-tooltip>
+        </div>
+        <Password v-model:value="awsCognitoIdentityId" placeholder="AWS Cognito Identity Pool ID" />
+      </div>
+       
+      <div class="section-item" v-show="voiceApiName=='Azure'">
         <label my-1 block for="">Azure 区域</label>
         <select
           v-model="azureRegion"
@@ -37,7 +64,7 @@ const { azureRegion, azureKey, isAlwaysRecognition, ttsPassword, autoPlay, voice
           </option>
         </select>
       </div>
-      <div class="section-item">
+      <div class="section-item" v-show="voiceApiName=='Azure'">
         <div center-y>
           <label mr-1 my-1 block for="">Azure Access Key</label>
           <el-tooltip