Initial commit: 军事科技AI配音系统 v1.0

- Flask backend with MiniMax and CosyVoice TTS engines
- Frontend UI with emotion tags, speed/pitch/volume controls
- Nginx reverse proxy configuration
- Dual engine support (MiniMax speech-2.8-hd, CosyVoice v3.5-flash)

backend/app/services/cosyvoice_service.py

@@ -0,0 +1,43 @@
+import os
+import requests
+
+def synthesize_cosyvoice(text, speed=1.0, pitch=0, volume=1.0, instruction=None):
+    """调用阿里云百炼CosyVoice API"""
+    api_key = os.getenv('DASHSCOPE_API_KEY', '')
+    if not api_key:
+        raise ValueError('DASHSCOPE_API_KEY未设置')
+    
+    url = 'https://dashscope.aliyuncs.com/api/v1/services/audio/tts/SpeechSynthesizer'
+    headers = {
+        'Authorization': f'Bearer {api_key}',
+        'Content-Type': 'application/json'
+    }
+    
+    payload = {
+        'model': 'cosyvoice-v3.5-flash',
+        'input': {
+            'text': text,
+            'voice': 'cosyvoice-v3.5-flash-bailian-e859a6e87823419f9c6b3eefb1a97dc3',
+            'format': 'mp3',
+            'sample_rate': 32000
+        }
+    }
+    
+    response = requests.post(url, json=payload, headers=headers, timeout=60)
+    
+    if response.status_code != 200:
+        raise Exception(f'CosyVoice API错误: {response.status_code} - {response.text}')
+    
+    result = response.json()
+    
+    # 阿里云返回JSON格式，音频数据在 output.audio.url 字段
+    if 'output' in result and 'audio' in result['output']:
+        audio_url = result['output']['audio'].get('url')
+        if audio_url:
+            # 从URL下载实际音频
+            audio_response = requests.get(audio_url, timeout=60)
+            if audio_response.status_code == 200:
+                return audio_response.content
+            raise Exception(f'下载音频失败: {audio_response.status_code}')
+    
+    raise Exception(f'CosyVoice返回格式异常: {result}')

backend/app/services/minimax_service.py

@@ -0,0 +1,50 @@
+import os
+import base64
+import requests
+from io import BytesIO
+
+def synthesize_minimax(text, speed=1.0, pitch=0, volume=1.0):
+    """调用MiniMax TTS API"""
+    api_key = os.getenv('MINIMAX_API_KEY', '')
+    if not api_key:
+        raise ValueError('MINIMAX_API_KEY未设置')
+    
+    # 将音量映射到0-100
+    volume_int = int(volume * 100)
+    
+    url = 'https://api.minimax.chat/v1/t2a_v2'
+    headers = {
+        'Authorization': f'Bearer {api_key}',
+        'Content-Type': 'application/json'
+    }
+    
+    payload = {
+        'model': 'speech-2.8-hd',
+        'text': text,
+        'stream': False,
+        'voice_setting': {
+            'voice_id': 'lanhao_military_tech',
+            'speed': speed,
+            'vol': volume_int,
+            'pitch': pitch,
+        },
+        'audio_setting': {
+            'audio_sample_rate': 32000,
+            'bitrate': 128000,
+            'format': 'mp3'
+        }
+    }
+    
+    response = requests.post(url, json=payload, headers=headers, timeout=60)
+    
+    if response.status_code != 200:
+        raise Exception(f'MiniMax API错误: {response.status_code} - {response.text}')
+    
+    result = response.json()
+    
+    if 'data' in result and 'audio' in result['data']:
+        audio_hex = result['data']['audio']
+        audio_bytes = bytes.fromhex(audio_hex)
+        return audio_bytes
+    
+    raise Exception(f'MiniMax返回格式异常: {result}')