feat(audio): 统一音量配置并修复原声音量默认值问题

引入AudioVolumeDefaults类集中管理音量配置，确保全局一致性
修复原声音量默认值为0.7以解决短剧解说模式问题
添加音量验证和详细日志便于调试

.gitignore

@@ -34,3 +34,6 @@ resource/srt/*.srt
 app/models/faster-whisper-large-v2/*
 app/models/faster-whisper-large-v3/*
 app/models/bert/*
+
+bug清单.md
+task.md

app/models/schema.py

@@ -1,6 +1,6 @@
 import warnings
 from enum import Enum
-from typing import Any, List, Optional
+from typing import Any, List, Optional, Union
 
 import pydantic
 from pydantic import BaseModel, Field
@@ -13,6 +13,24 @@ warnings.filterwarnings(
 )
 
 
+class AudioVolumeDefaults:
+    """音量配置默认值常量类 - 确保全局一致性"""
+
+    # 语音音量默认值
+    VOICE_VOLUME = 1.0
+    TTS_VOLUME = 1.0
+
+    # 原声音量默认值 - 这是修复bug的关键
+    ORIGINAL_VOLUME = 0.7
+
+    # 背景音乐音量默认值
+    BGM_VOLUME = 0.3
+
+    # 音量范围
+    MIN_VOLUME = 0.0
+    MAX_VOLUME = 1.0
+
+
 class VideoConcatMode(str, Enum):
     random = "random"
     sequential = "sequential"
@@ -101,7 +119,7 @@ class VideoParams(BaseModel):
 
     video_subject: str
     video_script: str = ""  # 用于生成视频的脚本
-    video_terms: Optional[str | list] = None  # 用于生成视频的关键词
+    video_terms: Optional[Union[str, list]] = None  # 用于生成视频的关键词
     video_aspect: Optional[VideoAspect] = VideoAspect.portrait.value
     video_concat_mode: Optional[VideoConcatMode] = VideoConcatMode.random.value
     video_clip_duration: Optional[int] = 5
@@ -113,11 +131,11 @@ class VideoParams(BaseModel):
     video_language: Optional[str] = ""  # auto detect
 
     voice_name: Optional[str] = ""
-    voice_volume: Optional[float] = 1.0
+    voice_volume: Optional[float] = AudioVolumeDefaults.VOICE_VOLUME
     voice_rate: Optional[float] = 1.0
     bgm_type: Optional[str] = "random"
     bgm_file: Optional[str] = ""
-    bgm_volume: Optional[float] = 0.2
+    bgm_volume: Optional[float] = AudioVolumeDefaults.BGM_VOLUME
 
     subtitle_enabled: Optional[bool] = True
     subtitle_position: Optional[str] = "bottom"  # top, bottom, center
@@ -157,11 +175,11 @@ class AudioRequest(BaseModel):
     video_script: str
     video_language: Optional[str] = ""
     voice_name: Optional[str] = "zh-CN-XiaoxiaoNeural-Female"
-    voice_volume: Optional[float] = 1.0
+    voice_volume: Optional[float] = AudioVolumeDefaults.VOICE_VOLUME
     voice_rate: Optional[float] = 1.2
     bgm_type: Optional[str] = "random"
     bgm_file: Optional[str] = ""
-    bgm_volume: Optional[float] = 0.2
+    bgm_volume: Optional[float] = AudioVolumeDefaults.BGM_VOLUME
     video_source: Optional[str] = "local"
 
 
@@ -347,7 +365,7 @@ class VideoClipParams(BaseModel):
     # video_concat_mode: Optional[VideoConcatMode] = VideoConcatMode.random.value
 
     voice_name: Optional[str] = Field(default="zh-CN-YunjianNeural", description="语音名称")
-    voice_volume: Optional[float] = Field(default=1.0, description="解说语音音量")
+    voice_volume: Optional[float] = Field(default=AudioVolumeDefaults.VOICE_VOLUME, description="解说语音音量")
     voice_rate: Optional[float] = Field(default=1.0, description="语速")
     voice_pitch: Optional[float] = Field(default=1.0, description="语调")
 
@@ -367,9 +385,9 @@ class VideoClipParams(BaseModel):
 
     n_threads: Optional[int] = Field(default=16, description="线程数")    # 线程数，有助于提升视频处理速度
 
-    tts_volume: Optional[float] = Field(default=1.0, description="解说语音音量（后处理）")
+    tts_volume: Optional[float] = Field(default=AudioVolumeDefaults.TTS_VOLUME, description="解说语音音量（后处理）")
-    original_volume: Optional[float] = Field(default=1.0, description="视频原声音量")
+    original_volume: Optional[float] = Field(default=AudioVolumeDefaults.ORIGINAL_VOLUME, description="视频原声音量")
-    bgm_volume: Optional[float] = Field(default=0.3, description="背景音乐音量")
+    bgm_volume: Optional[float] = Field(default=AudioVolumeDefaults.BGM_VOLUME, description="背景音乐音量")
 
 
 class VideoTranscriptionRequest(BaseModel):

app/services/generate_video.py

@@ -24,6 +24,7 @@ from moviepy.video.tools.subtitles import SubtitlesClip
 from PIL import ImageFont
 
 from app.utils import utils
+from app.models.schema import AudioVolumeDefaults
 
 
 def merge_materials(
@@ -66,11 +67,12 @@ def merge_materials(
     if options is None:
         options = {}
     
-    # 设置默认参数值
+    # 设置默认参数值 - 使用统一的音量配置
-    voice_volume = options.get('voice_volume', 1.0)
+    voice_volume = options.get('voice_volume', AudioVolumeDefaults.VOICE_VOLUME)
-    bgm_volume = options.get('bgm_volume', 0.3)
+    bgm_volume = options.get('bgm_volume', AudioVolumeDefaults.BGM_VOLUME)
-    original_audio_volume = options.get('original_audio_volume', 0.0)  # 默认为0，即不保留原声
+    # 修复bug: 将原声音量默认值从0.0改为0.7，确保短剧解说模式下原片音量正常
-    keep_original_audio = options.get('keep_original_audio', False)  # 是否保留原声
+    original_audio_volume = options.get('original_audio_volume', AudioVolumeDefaults.ORIGINAL_VOLUME)
+    keep_original_audio = options.get('keep_original_audio', True)  # 默认保留原声
     subtitle_font = options.get('subtitle_font', '')
     subtitle_font_size = options.get('subtitle_font_size', 40)
     subtitle_color = options.get('subtitle_color', '#FFFFFF')
@@ -82,6 +84,24 @@ def merge_materials(
     threads = options.get('threads', 2)
     fps = options.get('fps', 30)
 
+    # 音量配置日志 - 便于调试音量问题
+    logger.info(f"音量配置详情:")
+    logger.info(f"  - 配音音量: {voice_volume}")
+    logger.info(f"  - 背景音乐音量: {bgm_volume}")
+    logger.info(f"  - 原声音量: {original_audio_volume}")
+    logger.info(f"  - 是否保留原声: {keep_original_audio}")
+
+    # 音量参数验证
+    def validate_volume(volume, name):
+        if not (AudioVolumeDefaults.MIN_VOLUME <= volume <= AudioVolumeDefaults.MAX_VOLUME):
+            logger.warning(f"{name}音量 {volume} 超出有效范围 [{AudioVolumeDefaults.MIN_VOLUME}, {AudioVolumeDefaults.MAX_VOLUME}]，将被限制")
+            return max(AudioVolumeDefaults.MIN_VOLUME, min(volume, AudioVolumeDefaults.MAX_VOLUME))
+        return volume
+
+    voice_volume = validate_volume(voice_volume, "配音")
+    bgm_volume = validate_volume(bgm_volume, "背景音乐")
+    original_audio_volume = validate_volume(original_audio_volume, "原声")
+
     # 处理透明背景色问题 - MoviePy 2.1.1不支持'transparent'值
     if subtitle_bg_color == 'transparent':
         subtitle_bg_color = None  # None在新版MoviePy中表示透明背景

app/services/video.py

@@ -314,24 +314,35 @@ def generate_video_v3(
     audio_clips = []
 
     # 添加原声（设置音量）
-    logger.debug(f"音量配置: {volume_config}")
+    logger.info(f"音量配置详情: {volume_config}")
     if video.audio is not None:
-        original_audio = video.audio.volumex(volume_config['original'])
+        original_volume = volume_config['original']
+        logger.info(f"应用原声音量: {original_volume}")
+        original_audio = video.audio.volumex(original_volume)
         audio_clips.append(original_audio)
+        logger.info("原声音频已添加到合成列表")
+    else:
+        logger.warning("视频没有音轨，无法添加原声")
 
     # 添加BGM（如果提供）
     if bgm_path:
+        logger.info(f"添加背景音乐: {bgm_path}")
         bgm = AudioFileClip(bgm_path)
         if bgm.duration < video.duration:
             bgm = loop_audio_clip(bgm, video.duration)
         else:
             bgm = bgm.subclip(0, video.duration)
-        bgm = bgm.volumex(volume_config['bgm'])
+        bgm_volume = volume_config['bgm']
+        logger.info(f"应用BGM音量: {bgm_volume}")
+        bgm = bgm.volumex(bgm_volume)
         audio_clips.append(bgm)
 
     # 添加解说音频（如果提供）
     if narration_path:
-        narration = AudioFileClip(narration_path).volumex(volume_config['narration'])
+        logger.info(f"添加解说音频: {narration_path}")
+        narration_volume = volume_config['narration']
+        logger.info(f"应用解说音量: {narration_volume}")
+        narration = AudioFileClip(narration_path).volumex(narration_volume)
         audio_clips.append(narration)
 
     # 合成最终视频（包含字幕）
@@ -342,8 +353,12 @@ def generate_video_v3(
         final_video = video
 
     if audio_clips:
+        logger.info(f"合成音频轨道，共 {len(audio_clips)} 个音频片段")
         final_audio = CompositeAudioClip(audio_clips)
         final_video = final_video.set_audio(final_audio)
+        logger.info("音频合成完成")
+    else:
+        logger.warning("没有音频轨道需要合成")
 
     # 导出视频
     logger.info("开始导出视频...")  # 调试信息

webui/components/audio_settings.py

@@ -3,6 +3,7 @@ import os
 from uuid import uuid4
 from app.config import config
 from app.services import voice
+from app.models.schema import AudioVolumeDefaults
 from app.utils import utils
 from webui.utils.cache import get_songs_cache
 
@@ -94,12 +95,12 @@ def render_azure_v2_settings(tr):
 
 def render_voice_parameters(tr):
     """渲染语音参数设置"""
-    # 音量
+    # 音量 - 使用统一的默认值
     voice_volume = st.slider(
         tr("Speech Volume"),
-        min_value=0.0,
+        min_value=AudioVolumeDefaults.MIN_VOLUME,
-        max_value=1.0,
+        max_value=AudioVolumeDefaults.MAX_VOLUME,
-        value=1.0,
+        value=AudioVolumeDefaults.VOICE_VOLUME,
         step=0.01,
         help=tr("Adjust the volume of the original audio")
     )
@@ -187,12 +188,12 @@ def render_bgm_settings(tr):
         if custom_bgm_file and os.path.exists(custom_bgm_file):
             st.session_state['bgm_file'] = custom_bgm_file
 
-    # 背景音乐音量
+    # 背景音乐音量 - 使用统一的默认值
     bgm_volume = st.slider(
         tr("Background Music Volume"),
-        min_value=0.0,
+        min_value=AudioVolumeDefaults.MIN_VOLUME,
-        max_value=1.0,
+        max_value=AudioVolumeDefaults.MAX_VOLUME,
-        value=0.3,
+        value=AudioVolumeDefaults.BGM_VOLUME,
         step=0.01,
         help=tr("Adjust the volume of the original audio")
     )
@@ -203,10 +204,10 @@ def get_audio_params():
     """获取音频参数"""
     return {
         'voice_name': config.ui.get("voice_name", ""),
-        'voice_volume': st.session_state.get('voice_volume', 1.0),
+        'voice_volume': st.session_state.get('voice_volume', AudioVolumeDefaults.VOICE_VOLUME),
         'voice_rate': st.session_state.get('voice_rate', 1.0),
         'voice_pitch': st.session_state.get('voice_pitch', 1.0),
         'bgm_type': st.session_state.get('bgm_type', 'random'),
         'bgm_file': st.session_state.get('bgm_file', ''),
-        'bgm_volume': st.session_state.get('bgm_volume', 0.3),
+        'bgm_volume': st.session_state.get('bgm_volume', AudioVolumeDefaults.BGM_VOLUME),
     }

webui/components/video_settings.py

@@ -1,5 +1,5 @@
 import streamlit as st
-from app.models.schema import VideoClipParams, VideoAspect
+from app.models.schema import VideoClipParams, VideoAspect, AudioVolumeDefaults
 
 
 def render_video_panel(tr):
@@ -41,12 +41,12 @@ def render_video_config(tr, params):
     )
     st.session_state['video_quality'] = video_qualities[quality_index][1]
 
-    # 原声音量
+    # 原声音量 - 使用统一的默认值
     params.original_volume = st.slider(
         tr("Original Volume"),
-        min_value=0.0,
+        min_value=AudioVolumeDefaults.MIN_VOLUME,
-        max_value=1.0,
+        max_value=AudioVolumeDefaults.MAX_VOLUME,
-        value=0.7,
+        value=AudioVolumeDefaults.ORIGINAL_VOLUME,
         step=0.01,
         help=tr("Adjust the volume of the original audio")
     )
@@ -58,5 +58,5 @@ def get_video_params():
     return {
         'video_aspect': st.session_state.get('video_aspect', VideoAspect.portrait.value),
         'video_quality': st.session_state.get('video_quality', '1080p'),
-        'original_volume': st.session_state.get('original_volume', 0.7)
+        'original_volume': st.session_state.get('original_volume', AudioVolumeDefaults.ORIGINAL_VOLUME)
     }