feat: 完整的语音助手系统实现
主要功能: - ✅ 离线语音识别 (ASR) - Paraformer中文模型 - ✅ 在线语音识别 - Streaming Paraformer中英文双语模型 - ✅ 语音合成 (TTS) - MeloTTS中英文混合模型 - ✅ 语音唤醒 (KWS) - Zipformer关键词检测模型 - ✅ 麦克风录音功能 - 支持多种格式和实时转换 - ✅ 模型设置界面 - 完整的图形化配置管理 KWS优化亮点: - 🎯 成功实现关键词检测 (测试成功率10%→预期50%+) - ⚙️ 可调参数: 阈值、活跃路径、尾随空白、分数权重、线程数 - 🔧 智能参数验证和实时反馈 - 📊 详细的调试信息和成功统计 - 🎛️ 用户友好的设置界面 技术架构: - 模块化设计: ASRManager, TTSManager, KWSManager - 实时音频处理: 自动格式转换 (任意格式→16kHz单声道) - 智能设备检测: 自动选择最佳音频格式 - 完整资源管理: 正确的创建和销毁流程 - 跨平台支持: macOS优化的音频权限处理 界面特性: - 2×2网格布局: ASR、TTS、录音、KWS四大功能模块 - 分离录音设置: 设备参数 + 输出格式独立配置 - 实时状态显示: 音频电平、处理次数、成功统计 - 详细的用户指导和错误提示
This commit is contained in:
61
scripts/build.sh
Normal file
61
scripts/build.sh
Normal file
@@ -0,0 +1,61 @@
|
||||
#!/bin/bash
|
||||
|
||||
# QSmartAssistant 语音测试工具构建脚本
|
||||
|
||||
set -e # 遇到错误时退出
|
||||
|
||||
echo "=== QSmartAssistant 语音测试工具构建脚本 ==="
|
||||
|
||||
# 检查是否在正确的目录
|
||||
if [ ! -f "CMakeLists.txt" ]; then
|
||||
echo "错误: 请在项目根目录运行此脚本"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# 创建构建目录
|
||||
BUILD_DIR="build"
|
||||
if [ -d "$BUILD_DIR" ]; then
|
||||
echo "清理现有构建目录..."
|
||||
rm -rf "$BUILD_DIR"
|
||||
fi
|
||||
|
||||
echo "创建构建目录: $BUILD_DIR"
|
||||
mkdir -p "$BUILD_DIR"
|
||||
cd "$BUILD_DIR"
|
||||
|
||||
# 检查Qt6
|
||||
echo "检查Qt6安装..."
|
||||
if ! command -v qmake6 &> /dev/null && ! command -v qmake &> /dev/null; then
|
||||
echo "警告: 未找到Qt6，请确保已正确安装Qt6"
|
||||
fi
|
||||
|
||||
# 配置CMake
|
||||
echo "配置CMake..."
|
||||
if [ -n "$SHERPA_ONNX_ROOT" ]; then
|
||||
echo "使用自定义sherpa-onnx路径: $SHERPA_ONNX_ROOT"
|
||||
cmake -DSHERPA_ONNX_ROOT="$SHERPA_ONNX_ROOT" ..
|
||||
else
|
||||
echo "使用默认sherpa-onnx路径"
|
||||
cmake ..
|
||||
fi
|
||||
|
||||
# 编译
|
||||
echo "开始编译..."
|
||||
CPU_COUNT=$(sysctl -n hw.ncpu 2>/dev/null || nproc 2>/dev/null || echo 4)
|
||||
make -j$CPU_COUNT
|
||||
|
||||
# 检查编译结果
|
||||
if [ -f "qt_speech_simple" ]; then
|
||||
echo "=== 编译成功! ==="
|
||||
echo "可执行文件位置: $(pwd)/qt_speech_simple"
|
||||
echo ""
|
||||
echo "运行程序:"
|
||||
echo " cd $(pwd)"
|
||||
echo " ./qt_speech_simple"
|
||||
echo ""
|
||||
echo "注意: 请确保模型文件已正确放置在 ~/.config/QSmartAssistant/Data/ 目录下"
|
||||
else
|
||||
echo "=== 编译失败! ==="
|
||||
echo "请检查错误信息并解决依赖问题"
|
||||
exit 1
|
||||
fi
|
||||
123
scripts/check_audio_permissions.sh
Executable file
123
scripts/check_audio_permissions.sh
Executable file
@@ -0,0 +1,123 @@
|
||||
#!/bin/bash
|
||||
|
||||
echo "=== macOS 麦克风权限诊断和修复工具 ==="
|
||||
echo "当前时间: $(date)"
|
||||
echo "用户: $(whoami)"
|
||||
echo ""
|
||||
|
||||
# 1. 检查音频设备
|
||||
echo "📱 1. 音频设备检查"
|
||||
echo "----------------------------------------"
|
||||
system_profiler SPAudioDataType | grep -E "(MacBook Pro|Built-in|Microphone)" || echo "未找到内置麦克风设备"
|
||||
echo ""
|
||||
|
||||
# 2. 检查麦克风权限状态
|
||||
echo "🔐 2. 麦克风权限状态检查"
|
||||
echo "----------------------------------------"
|
||||
# 尝试读取TCC数据库
|
||||
if [ -f ~/Library/Application\ Support/com.apple.TCC/TCC.db ]; then
|
||||
echo "TCC数据库存在，检查权限记录..."
|
||||
sqlite3 ~/Library/Application\ Support/com.apple.TCC/TCC.db \
|
||||
"SELECT client, auth_value, auth_reason FROM access WHERE service='kTCCServiceMicrophone';" 2>/dev/null | \
|
||||
while IFS='|' read -r client auth_value auth_reason; do
|
||||
if [ ! -z "$client" ]; then
|
||||
status="未知"
|
||||
case $auth_value in
|
||||
0) status="拒绝" ;;
|
||||
1) status="允许" ;;
|
||||
2) status="允许" ;;
|
||||
3) status="限制" ;;
|
||||
esac
|
||||
echo "应用: $client -> 权限: $status ($auth_value)"
|
||||
fi
|
||||
done
|
||||
else
|
||||
echo "TCC数据库不存在或无法访问"
|
||||
fi
|
||||
echo ""
|
||||
|
||||
# 3. 测试系统音频录制能力
|
||||
echo "🎤 3. 系统音频录制测试"
|
||||
echo "----------------------------------------"
|
||||
if command -v sox >/dev/null 2>&1; then
|
||||
echo "使用sox进行录制测试..."
|
||||
timeout 2s rec -q -t wav /tmp/test_audio_$(date +%s).wav 2>/dev/null
|
||||
if [ $? -eq 0 ] && [ -f /tmp/test_audio_*.wav ]; then
|
||||
audio_file=$(ls /tmp/test_audio_*.wav | head -1)
|
||||
file_size=$(stat -f%z "$audio_file" 2>/dev/null || echo "0")
|
||||
echo "✅ 录制成功！文件大小: ${file_size} 字节"
|
||||
rm -f /tmp/test_audio_*.wav
|
||||
else
|
||||
echo "❌ 录制失败 - 可能是权限问题"
|
||||
fi
|
||||
else
|
||||
echo "⚠️ sox未安装，跳过录制测试"
|
||||
echo " 可以通过 'brew install sox' 安装"
|
||||
fi
|
||||
echo ""
|
||||
|
||||
# 4. 检查Qt程序的权限状态
|
||||
echo "🖥️ 4. Qt程序权限检查"
|
||||
echo "----------------------------------------"
|
||||
qt_app_path="./cmake-build-debug/qt_speech_simple"
|
||||
if [ -f "$qt_app_path" ]; then
|
||||
echo "Qt程序路径: $qt_app_path"
|
||||
|
||||
# 检查程序是否在TCC数据库中
|
||||
app_bundle_id=$(basename "$qt_app_path")
|
||||
echo "检查程序ID: $app_bundle_id"
|
||||
|
||||
# 尝试查找相关权限记录
|
||||
sqlite3 ~/Library/Application\ Support/com.apple.TCC/TCC.db \
|
||||
"SELECT client, auth_value FROM access WHERE service='kTCCServiceMicrophone' AND client LIKE '%$app_bundle_id%';" 2>/dev/null | \
|
||||
while IFS='|' read -r client auth_value; do
|
||||
if [ ! -z "$client" ]; then
|
||||
status="拒绝"
|
||||
[ "$auth_value" = "2" ] && status="允许"
|
||||
echo "找到权限记录: $client -> $status"
|
||||
fi
|
||||
done
|
||||
else
|
||||
echo "❌ Qt程序不存在: $qt_app_path"
|
||||
fi
|
||||
echo ""
|
||||
|
||||
# 5. 权限修复建议
|
||||
echo "🔧 5. 权限修复步骤"
|
||||
echo "----------------------------------------"
|
||||
echo "如果遇到权限问题，请按以下步骤操作："
|
||||
echo ""
|
||||
echo "方法1: 通过系统设置授予权限"
|
||||
echo " 1. 打开 系统设置 (System Preferences)"
|
||||
echo " 2. 点击 安全性与隐私 (Security & Privacy)"
|
||||
echo " 3. 选择 隐私 (Privacy) 标签"
|
||||
echo " 4. 在左侧列表中选择 麦克风 (Microphone)"
|
||||
echo " 5. 确保Qt程序已勾选并允许访问麦克风"
|
||||
echo ""
|
||||
echo "方法2: 重置麦克风权限 (需要管理员权限)"
|
||||
echo " sudo tccutil reset Microphone"
|
||||
echo " 然后重新运行Qt程序，会再次弹出权限请求"
|
||||
echo ""
|
||||
echo "方法3: 手动添加权限 (macOS Monterey及以上)"
|
||||
echo " 1. 系统设置 -> 隐私与安全性 -> 麦克风"
|
||||
echo " 2. 点击 + 号添加应用程序"
|
||||
echo " 3. 选择Qt程序可执行文件"
|
||||
echo ""
|
||||
|
||||
# 6. 启动Qt程序进行实际测试
|
||||
echo "🚀 6. 启动Qt程序测试"
|
||||
echo "----------------------------------------"
|
||||
if [ -f "$qt_app_path" ]; then
|
||||
echo "即将启动Qt程序进行麦克风权限测试..."
|
||||
echo "请注意观察是否弹出权限请求对话框"
|
||||
echo "如果弹出，请点击 '允许' 或 'Allow'"
|
||||
echo ""
|
||||
echo "按回车键继续启动程序，或Ctrl+C取消..."
|
||||
read -r
|
||||
|
||||
echo "启动程序: $qt_app_path"
|
||||
cd cmake-build-debug && ./qt_speech_simple
|
||||
else
|
||||
echo "❌ 程序文件不存在，请先编译项目"
|
||||
echo "运行: mkdir -p cmake-build-debug && cd cmake-build-debug && cmake .. && make"
|
||||
fi
|
||||
126
scripts/fix_microphone_permission.sh
Executable file
126
scripts/fix_microphone_permission.sh
Executable file
@@ -0,0 +1,126 @@
|
||||
#!/bin/bash
|
||||
|
||||
# 快速麦克风权限修复脚本
|
||||
# 用于解决macOS上Qt程序的麦克风权限问题
|
||||
|
||||
set -e
|
||||
|
||||
echo "🎤 Qt语音识别程序 - 麦克风权限快速修复"
|
||||
echo "============================================"
|
||||
echo ""
|
||||
|
||||
# 检查是否为macOS系统
|
||||
if [[ "$OSTYPE" != "darwin"* ]]; then
|
||||
echo "❌ 此脚本仅适用于macOS系统"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# 检查程序文件
|
||||
QT_APP="./cmake-build-debug/qt_speech_simple"
|
||||
if [ ! -f "$QT_APP" ]; then
|
||||
echo "❌ Qt程序不存在: $QT_APP"
|
||||
echo "请先编译项目："
|
||||
echo " mkdir -p cmake-build-debug"
|
||||
echo " cd cmake-build-debug"
|
||||
echo " cmake .."
|
||||
echo " make"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
echo "✅ 找到Qt程序: $QT_APP"
|
||||
echo ""
|
||||
|
||||
# 显示当前权限状态
|
||||
echo "📋 当前麦克风权限状态："
|
||||
echo "----------------------------------------"
|
||||
if sqlite3 ~/Library/Application\ Support/com.apple.TCC/TCC.db \
|
||||
"SELECT client, auth_value FROM access WHERE service='kTCCServiceMicrophone';" 2>/dev/null | grep -q "qt_speech_simple"; then
|
||||
echo "✅ 找到程序的权限记录"
|
||||
else
|
||||
echo "⚠️ 未找到程序的权限记录"
|
||||
fi
|
||||
echo ""
|
||||
|
||||
# 提供修复选项
|
||||
echo "🔧 请选择修复方法："
|
||||
echo "----------------------------------------"
|
||||
echo "1. 重置所有麦克风权限（推荐）"
|
||||
echo "2. 打开系统设置手动配置"
|
||||
echo "3. 直接启动程序测试"
|
||||
echo "4. 退出"
|
||||
echo ""
|
||||
|
||||
read -p "请输入选择 (1-4): " choice
|
||||
|
||||
case $choice in
|
||||
1)
|
||||
echo ""
|
||||
echo "🔄 重置麦克风权限..."
|
||||
if sudo tccutil reset Microphone; then
|
||||
echo "✅ 权限重置成功"
|
||||
echo ""
|
||||
echo "📱 即将启动Qt程序，请注意："
|
||||
echo " 1. 程序启动时会弹出权限请求对话框"
|
||||
echo " 2. 请点击 '允许' 或 'Allow'"
|
||||
echo " 3. 如果没有弹出对话框，请手动在系统设置中添加权限"
|
||||
echo ""
|
||||
read -p "按回车键启动程序..."
|
||||
|
||||
# 重启音频服务确保权限生效
|
||||
echo "🔄 重启音频服务..."
|
||||
sudo killall coreaudiod 2>/dev/null || true
|
||||
sleep 2
|
||||
|
||||
# 启动程序
|
||||
echo "🚀 启动Qt程序..."
|
||||
cd cmake-build-debug
|
||||
./qt_speech_simple
|
||||
else
|
||||
echo "❌ 权限重置失败，可能需要管理员权限"
|
||||
fi
|
||||
;;
|
||||
2)
|
||||
echo ""
|
||||
echo "📱 打开系统设置进行手动配置..."
|
||||
echo ""
|
||||
echo "请按以下步骤操作："
|
||||
echo "1. 系统设置 → 隐私与安全性 → 麦克风"
|
||||
echo "2. 点击右侧的 + 按钮"
|
||||
echo "3. 浏览到: $(pwd)/cmake-build-debug/qt_speech_simple"
|
||||
echo "4. 选择程序并确保开关为开启状态"
|
||||
echo ""
|
||||
|
||||
# 尝试打开系统设置
|
||||
open "x-apple.systempreferences:com.apple.preference.security?Privacy_Microphone" 2>/dev/null || \
|
||||
open "/System/Library/PreferencePanes/Security.prefPane" 2>/dev/null || \
|
||||
echo "请手动打开系统设置"
|
||||
|
||||
read -p "配置完成后按回车键启动程序..."
|
||||
cd cmake-build-debug
|
||||
./qt_speech_simple
|
||||
;;
|
||||
3)
|
||||
echo ""
|
||||
echo "🚀 直接启动程序进行测试..."
|
||||
cd cmake-build-debug
|
||||
./qt_speech_simple
|
||||
;;
|
||||
4)
|
||||
echo "👋 退出脚本"
|
||||
exit 0
|
||||
;;
|
||||
*)
|
||||
echo "❌ 无效选择"
|
||||
exit 1
|
||||
;;
|
||||
esac
|
||||
|
||||
echo ""
|
||||
echo "🎉 脚本执行完成！"
|
||||
echo ""
|
||||
echo "💡 如果仍有问题，请查看详细文档："
|
||||
echo " - docs/MICROPHONE_PERMISSION_FIX.md"
|
||||
echo " - docs/MICROPHONE_RECOGNITION_GUIDE.md"
|
||||
echo ""
|
||||
echo "或运行完整诊断脚本："
|
||||
echo " ./check_audio_permissions.sh"
|
||||
Reference in New Issue
Block a user