diff --git a/backend/app/routes/tts_synthesize.py b/backend/app/routes/tts_synthesize.py
index e4b2c1b..348af9f 100644
--- a/backend/app/routes/tts_synthesize.py
+++ b/backend/app/routes/tts_synthesize.py
@@ -4,6 +4,7 @@ import os
 from app.services.minimax_service import synthesize_minimax
 from app.services.cosyvoice_service import synthesize_cosyvoice
 from app.services.doubao_service import synthesize_doubao
+from app.services.fish_audio_service import synthesize_fish_audio
 from app.routes.auth import login_required
 from app.db import get_db
 
@@ -43,6 +44,11 @@ def synthesize():
                 text=text, speed=speed, pitch=pitch,
                 volume=volume, emotion=emotion
             )
+        elif engine == 'fish_audio':
+            audio_data = synthesize_fish_audio(
+                text=text, speed=speed, pitch=pitch,
+                volume=volume, emotion=emotion
+            )
         else:
             audio_data = synthesize_minimax(
                 text=text, speed=speed, pitch=pitch,
diff --git a/backend/app/services/fish_audio_service.py b/backend/app/services/fish_audio_service.py
new file mode 100644
index 0000000..5bde26d
--- /dev/null
+++ b/backend/app/services/fish_audio_service.py
@@ -0,0 +1,46 @@
+import os
+import requests
+
+
+def synthesize_fish_audio(text, speed=1.0, pitch=0, volume=1.0, emotion='neutral'):
+    api_key = os.getenv('FISH_AUDIO_API_KEY', '')
+    if not api_key:
+        raise ValueError('FISH_AUDIO_API_KEY未设置')
+
+    voice_id = os.getenv('FISH_AUDIO_VOICE_ID', '34e841f0f4ff44aca6e396debf2776d4')
+
+    emotion_tags = {
+        'happy': '[happy]',
+        'sad': '[sad]',
+        'angry': '[angry]',
+        'surprised': '[excited]',
+        'fearful': '[nervous]',
+    }
+    tag = emotion_tags.get(emotion, '')
+    if tag:
+        text = f"{tag}{text}"
+
+    url = 'https://api.fish.audio/v1/tts'
+    headers = {
+        'Authorization': f'Bearer {api_key}',
+        'Content-Type': 'application/json'
+    }
+
+    payload = {
+        'text': text,
+        'reference_id': voice_id,
+        'format': 'mp3',
+        'speed': max(0.5, min(2.0, speed)),
+        'volume': max(-20, min(20, (volume - 1.0) * 20)),
+    }
+
+    response = requests.post(url, json=payload, headers=headers, timeout=60, stream=True)
+
+    if response.status_code != 200:
+        raise Exception(f'Fish Audio API错误: {response.status_code} - {response.text[:200]}')
+
+    audio_bytes = response.content
+    if not audio_bytes:
+        raise Exception('Fish Audio返回空音频')
+
+    return audio_bytes
diff --git a/frontend/app.js b/frontend/app.js
index b9e9cca..4e898bd 100644
--- a/frontend/app.js
+++ b/frontend/app.js
@@ -22,6 +22,11 @@ const ENGINE_CONFIG = {
         name: '豆包',
         emotionMode: 'global',
         color: '#7c5cfc'
+    },
+    fish_audio: {
+        name: 'Fish Audio',
+        emotionMode: 'global',
+        color: '#f59e0b'
     }
 };
 
diff --git a/frontend/index.html b/frontend/index.html
index 38ccadc..0479fff 100644
--- a/frontend/index.html
+++ b/frontend/index.html
@@ -41,6 +41,7 @@
                             <option value="minimax">MiniMax</option>
                             <option value="cosyvoice">CosyVoice</option>
                             <option value="doubao">豆包</option>
+                            <option value="fish_audio">Fish Audio</option>
                         </select>
                         <span class="toolbar-divider"></span>
                         <span class="toolbar-label">情绪：</span>