diff --git a/apps/application/serializers/application_serializers.py b/apps/application/serializers/application_serializers.py
index 9faabd4ef7..629d118a91 100644
--- a/apps/application/serializers/application_serializers.py
+++ b/apps/application/serializers/application_serializers.py
@@ -47,6 +47,7 @@
 from setting.models.model_management import Model
 from setting.models_provider import get_model_credential
 from setting.models_provider.constants.model_provider_constants import ModelProvideConstants
+from setting.models_provider.tools import get_model_instance_by_model_user_id
 from setting.serializers.provider_serializers import ModelSerializer
 from smartdoc.conf import PROJECT_DIR
 
@@ -856,15 +857,24 @@ def get_work_flow_model(instance):
                     instance['tts_model_enable'] = node['properties']['node_data']['tts_model_enable']
                     break
 
-        def speech_to_text(self, filelist):
-            # todo 找到模型 mp3转text
-            print(self.application_id)
-            print(filelist)
+        def speech_to_text(self, file, with_valid=True):
+            if with_valid:
+                self.is_valid(raise_exception=True)
+            application_id = self.data.get('application_id')
+            application = QuerySet(Application).filter(id=application_id).first()
+            if application.stt_model_enable:
+                model = get_model_instance_by_model_user_id(application.stt_model_id, application.user_id)
+                text = model.speech_to_text(file)
+                return text
 
-        def text_to_speech(self, text):
-            # todo 找到模型 text转bytes
-            print(self.application_id)
-            print(text)
+        def text_to_speech(self, text, with_valid=True):
+            if with_valid:
+                self.is_valid(raise_exception=True)
+            application_id = self.data.get('application_id')
+            application = QuerySet(Application).filter(id=application_id).first()
+            if application.tts_model_enable:
+                model = get_model_instance_by_model_user_id(application.tts_model_id, application.user_id)
+                return model.text_to_speech(text)
 
     class ApplicationKeySerializerModel(serializers.ModelSerializer):
         class Meta:
diff --git a/apps/application/urls.py b/apps/application/urls.py
index 2b7e779fae..5a41bc59ce 100644
--- a/apps/application/urls.py
+++ b/apps/application/urls.py
@@ -64,7 +64,7 @@
         'application/<str:application_id>/chat/<chat_id>/chat_record/<str:chat_record_id>/dataset/<str:dataset_id>/document_id/<str:document_id>/improve/<str:paragraph_id>',
         views.ChatView.ChatRecord.Improve.Operate.as_view(),
         name=''),
-    path('application/<str:application_id>/<str:model_id>/speech_to_text', views.Application.SpeechToText.as_view(), name='application/audio'),
-    path('application/<str:application_id>/<str:model_id>/text_to_speech', views.Application.TextToSpeech.as_view(), name='application/audio'),
+    path('application/<str:application_id>/speech_to_text', views.Application.SpeechToText.as_view(), name='application/audio'),
+    path('application/<str:application_id>/text_to_speech', views.Application.TextToSpeech.as_view(), name='application/audio'),
 
 ]
diff --git a/apps/application/views/application_views.py b/apps/application/views/application_views.py
index db0dae034c..a9dd2fe4dd 100644
--- a/apps/application/views/application_views.py
+++ b/apps/application/views/application_views.py
@@ -545,10 +545,9 @@ class SpeechToText(APIView):
                                                                         dynamic_tag=keywords.get(
                                                                             'application_id'))],
                                         compare=CompareConstants.AND))
-        def post(self, request: Request, application_id: str, model_id: str):
+        def post(self, request: Request, application_id: str):
             return result.success(
-                ApplicationSerializer.Operate(
-                    data={'application_id': application_id, 'user_id': request.user.id, 'model_id': model_id})
+                ApplicationSerializer.Operate(data={'application_id': application_id, 'user_id': request.user.id})
                 .speech_to_text(request.FILES.getlist('file')[0]))
 
     class TextToSpeech(APIView):
@@ -561,8 +560,9 @@ class TextToSpeech(APIView):
                                                                         dynamic_tag=keywords.get(
                                                                             'application_id'))],
                                         compare=CompareConstants.AND))
-        def post(self, request: Request, application_id: str, model_id: str):
-            return result.success(
-                ApplicationSerializer.Operate(
-                    data={'application_id': application_id, 'user_id': request.user.id, 'model_id': model_id})
-                .text_to_speech(request.data.get('text')))
+        def post(self, request: Request, application_id: str):
+            byte_data = ApplicationSerializer.Operate(
+                data={'application_id': application_id, 'user_id': request.user.id}).text_to_speech(
+                request.data.get('text'))
+            return HttpResponse(byte_data, status=200, headers={'Content-Type': 'audio/mp3',
+                                                              'Content-Disposition': 'attachment; filename="abc.mp3"'})
diff --git a/ui/package.json b/ui/package.json
index 0a3635c206..3e1582bae3 100644
--- a/ui/package.json
+++ b/ui/package.json
@@ -43,7 +43,8 @@
     "vue-clipboard3": "^2.0.0",
     "vue-codemirror": "^6.1.1",
     "vue-i18n": "^9.13.1",
-    "vue-router": "^4.2.4"
+    "vue-router": "^4.2.4",
+    "recorder-core": "^1.3.24040900"
   },
   "devDependencies": {
     "@rushstack/eslint-patch": "^1.3.2",
diff --git a/ui/src/api/application.ts b/ui/src/api/application.ts
index 2cf3b413eb..f21c465b0f 100644
--- a/ui/src/api/application.ts
+++ b/ui/src/api/application.ts
@@ -1,5 +1,5 @@
 import { Result } from '@/request/Result'
-import { get, post, postStream, del, put } from '@/request/index'
+import { get, post, postStream, del, put, request, download } from '@/request/index'
 import type { pageRequest } from '@/api/type/common'
 import type { ApplicationFormType } from '@/api/type/application'
 import { type Ref } from 'vue'
@@ -330,6 +330,29 @@ const getModelParamsForm: (
 ) => Promise<Result<Array<FormField>>> = (application_id, model_id, loading) => {
   return get(`${prefix}/${application_id}/model_params_form/${model_id}`, undefined, loading)
 }
+
+/**
+ * 语音转文本
+ */
+const postSpeechToText: (
+  application_id: String,
+  data: any,
+  loading?: Ref<boolean>
+) => Promise<Result<any>> = (application_id, data, loading) => {
+  return post(`${prefix}/${application_id}/speech_to_text`, data, undefined, loading)
+}
+
+/**
+ * 语音转文本
+ */
+const postTextToSpeech: (
+  application_id: String,
+  data: any,
+  loading?: Ref<boolean>
+) => Promise<Result<any>> = (application_id, data, loading) => {
+  return download(`${prefix}/${application_id}/text_to_speech`, 'post', data, undefined, loading)
+}
+
 export default {
   getAllAppilcation,
   getApplication,
@@ -356,4 +379,6 @@ export default {
   getApplicationRerankerModel,
   getApplicationSTTModel,
   getApplicationTTSModel,
+  postSpeechToText,
+  postTextToSpeech,
 }
diff --git a/ui/src/components/ai-chat/index.vue b/ui/src/components/ai-chat/index.vue
index d72d13e793..074cb86d2b 100644
--- a/ui/src/components/ai-chat/index.vue
+++ b/ui/src/components/ai-chat/index.vue
@@ -114,6 +114,11 @@
                   @regeneration="regenerationChart(item)"
                 />
               </div>
+              <div style="float: right;" v-if="props.data.tts_model_enable">
+                <el-button :disabled="!item.write_ed" @click="playAnswerText(item.answer_text)">
+                  <el-icon><VideoPlay /></el-icon>
+                </el-button>
+              </div>
             </div>
           </div>
         </template>
@@ -131,6 +136,20 @@
           :maxlength="100000"
           @keydown.enter="sendChatHandle($event)"
         />
+        <div class="operate" v-if="props.data.stt_model_enable">
+          <el-button
+            v-if="mediaRecorderStatus"
+            @click="startRecording"
+          >
+            <el-icon><Microphone /></el-icon>
+          </el-button>
+          <el-button
+            v-else
+            @click="stopRecording"
+          >
+            <el-icon><VideoPause /></el-icon>
+          </el-button>
+        </div>
         <div class="operate">
           <el-button
             text
@@ -149,6 +168,8 @@
         </div>
       </div>
     </div>
+    <!-- 先渲染，不然不能播放   -->
+    <audio ref="audioPlayer" controls hidden="hidden"></audio>
   </div>
 </template>
 <script setup lang="ts">
@@ -165,6 +186,10 @@ import useStore from '@/stores'
 import MdRenderer from '@/components/markdown/MdRenderer.vue'
 import { isWorkFlow } from '@/utils/application'
 import { debounce } from 'lodash'
+import Recorder from 'recorder-core'
+import 'recorder-core/src/engine/mp3'
+import 'recorder-core/src/engine/mp3-engine'
+
 defineOptions({ name: 'AiChat' })
 const route = useRoute()
 const {
@@ -592,6 +617,103 @@ const handleScroll = () => {
   }
 }
 
+// 定义响应式引用
+const mediaRecorder= ref<any>(null)
+const audioPlayer= ref<HTMLAudioElement | null>(null)
+const mediaRecorderStatus = ref(true)
+
+
+// 开始录音
+const startRecording = async () => {
+  try {
+    mediaRecorderStatus.value = false
+    mediaRecorder.value = new Recorder({
+      type: 'mp3',
+      bitRate: 128,
+      sampleRate: 44100,
+    })
+
+    mediaRecorder.value.open(() => {
+        mediaRecorder.value.start()
+    }, (err: any) => {
+      console.error(err)
+    })
+  } catch (error) {
+    console.error('无法获取音频权限：', error)
+  }
+}
+
+// 停止录音
+const stopRecording = () => {
+  if (mediaRecorder.value) {
+    mediaRecorderStatus.value = true
+    mediaRecorder.value.stop((blob: Blob, duration: number) => {
+       // 测试blob是否能正常播放
+       //  const link = document.createElement('a')
+       //  link.href = window.URL.createObjectURL(blob)
+       //  link.download = 'abc.mp3'
+       //  link.click()
+
+        uploadRecording(blob) // 上传录音文件
+    }, (err: any) => {
+      console.error('录音失败:', err)
+    })
+  }
+}
+
+// 上传录音文件
+const uploadRecording = async (audioBlob: Blob) => {
+  try {
+    const formData = new FormData()
+    formData.append('file', audioBlob, 'recording.mp3')
+
+      if (id) {
+        applicationApi.postSpeechToText(id as string, formData, loading)
+          .then((response) => {
+            console.log('上传成功:', response.data)
+            inputValue.value = response.data
+            // chatMessage(null, res.data)
+          })
+      }
+
+  } catch (error) {
+    console.error('上传失败:', error)
+  }
+}
+
+const playAnswerText = (text: string) => {
+  if (id) {
+    console.log(text)
+    applicationApi.postTextToSpeech(id as string, { 'text': text }, loading)
+      .then((res: any) => {
+
+        // 假设我们有一个 MP3 文件的字节数组
+        // 创建 Blob 对象
+        const blob = new Blob([res], { type: 'audio/mp3' })
+
+        // 创建对象 URL
+        const url = URL.createObjectURL(blob)
+
+        // 测试blob是否能正常播放
+        // const link = document.createElement('a')
+        // link.href = window.URL.createObjectURL(blob)
+        // link.download = "abc.mp3"
+        // link.click()
+
+        // 检查 audioPlayer 是否已经引用了 DOM 元素
+        if (audioPlayer.value instanceof HTMLAudioElement) {
+          audioPlayer.value.src = url;
+          audioPlayer.value.play(); // 自动播放音频
+        } else {
+          console.error("audioPlayer.value is not an instance of HTMLAudioElement");
+        }
+      })
+      .catch((err) => {
+        console.log('err: ', err)
+      })
+  }
+}
+
 function setScrollBottom() {
   // 将滚动条滚动到最下面
   scrollDiv.value.setScrollTop(getMaxHeight())
diff --git a/ui/src/request/index.ts b/ui/src/request/index.ts
index f578863bd6..fdde51396a 100644
--- a/ui/src/request/index.ts
+++ b/ui/src/request/index.ts
@@ -240,6 +240,23 @@ export const exportExcel: (
     .catch((e) => {})
 }
 
+
+export const download: (
+  url: string,
+  method: string,
+  data?: any,
+  params?: any,
+  loading?: NProgress | Ref<boolean>
+) => Promise<any> = (
+  url: string,
+  method: string,
+  data?: any,
+  params?: any,
+  loading?: NProgress | Ref<boolean>
+) => {
+  return promise(request({ url: url, method: method, data, params, responseType: 'blob' }), loading)
+}
+
 /**
  * 与服务器建立ws链接
  * @param url websocket路径