Digital_Human/run_train_and_clone.sh

# 启用bash的扩展功能（对应bat的enabledelayedexpansion）
shopt -s extglob

# 检查参数是否传入
if [ $# -eq 0 ]; then
    echo "Error: Insufficient parameters"
    echo
    echo "Usage:"
    echo "  ./run_train_and_clone.sh \"source_image\""
    echo
    echo "Examples:"
    echo "  ./run_train_and_clone.sh \"LivePortrait-main/assets/examples/source/s9.jpg\""
    exit 1
fi

# 定义核心变量
SOURCE_IMAGE="$1"
DRIVING_VIDEO="LivePortrait-main/assets/examples/driving/output.mp4"

# 获取脚本所在目录（对应bat的%~dp0）
SCRIPT_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)/
cd "$SCRIPT_DIR" || exit 1

# 生成时间戳（对应bat的powershell时间戳）
TIMESTAMP=$(date +%Y%m%d_%H%M%S)
AVATAR_ID="avatar_${TIMESTAMP}"
AVATAR_DIR="${SCRIPT_DIR}data/avatars/${AVATAR_ID}"

# 停止脚本时自动删除本次生成的头像目录
cleanup_avatar_dir() {
    if [ -n "${AVATAR_DIR:-}" ] && [ -d "${AVATAR_DIR}" ]; then
        echo "Cleaning generated avatar directory: ${AVATAR_DIR}"
        rm -rf -- "${AVATAR_DIR}"
    fi
}
trap 'cleanup_avatar_dir' EXIT

# 默认环境变量（可通过外部传入覆盖）
: "${CUDA_VISIBLE_DEVICES:=6,7}"
: "${AIOICE_PORT_MIN:=50000}"
: "${AIOICE_PORT_MAX:=50010}"
: "${AIOICE_BIND_IP:=192.168.22.9}"
: "${WEBRTC_NAT_IP:=183.252.196.135}"
export CUDA_VISIBLE_DEVICES AIOICE_PORT_MIN AIOICE_PORT_MAX AIOICE_BIND_IP WEBRTC_NAT_IP

echo "========================================"
echo "Digital Human Generation Pipeline"
echo "========================================"
echo
echo "Source Image: ${SOURCE_IMAGE}"
echo "Driving Video: ${DRIVING_VIDEO} (Fixed)"
echo "Avatar ID: ${AVATAR_ID}"
echo "Timestamp: ${TIMESTAMP}"
echo "Using env: CUDA_VISIBLE_DEVICES=${CUDA_VISIBLE_DEVICES} AIOICE_BIND_IP=${AIOICE_BIND_IP} WEBRTC_NAT_IP=${WEBRTC_NAT_IP} AIOICE_PORT_MIN=${AIOICE_PORT_MIN} AIOICE_PORT_MAX=${AIOICE_PORT_MAX}"
echo

# 检查源图片是否存在
if [ ! -f "${SOURCE_IMAGE}" ]; then
    echo "Error: Source image not found: ${SOURCE_IMAGE}"
    exit 1
fi

# 检查驱动视频是否存在
if [ ! -f "${DRIVING_VIDEO}" ]; then
    echo "Error: Driving video not found: ${DRIVING_VIDEO}"
    exit 1
fi

echo "Current Directory: $(pwd)"
echo

# ========================================
# Step 1: Run LivePortrait to generate video
# ========================================
echo
echo "[Step 1/3] Running LivePortrait to generate video..."
echo

# 转换驱动视频为绝对路径（对应bat的DRIVING_VIDEO_ABS）
DRIVING_VIDEO_ABS="${DRIVING_VIDEO}"
if [[ ! "${DRIVING_VIDEO_ABS}" =~ ^/ ]]; then
    DRIVING_VIDEO_ABS="${SCRIPT_DIR}${DRIVING_VIDEO}"
fi

echo "Debug: SCRIPT_DIR = ${SCRIPT_DIR}"
echo "Debug: DRIVING_VIDEO_ABS = ${DRIVING_VIDEO_ABS}"
echo "Debug: File exists check:"

# 检查驱动视频绝对路径是否存在
if [ ! -f "${DRIVING_VIDEO_ABS}" ]; then
    echo "Error: Driving video file does not exist at: ${DRIVING_VIDEO_ABS}"
    echo
    echo "Checking if the path is correct..."
    echo "Checking: LivePortrait-main/assets/examples/driving/output.mp4"
    if [ -f "LivePortrait-main/assets/examples/driving/output.mp4" ]; then
        echo "File exists with relative path!"
    else
        echo "File NOT found with relative path!"
    fi
    exit 1
fi

# 创建临时驱动视频副本（对应bat的TEMP_DRIVING）
DRIVE_EXT="${DRIVING_VIDEO_ABS##*.}"
DRIVE_BASE="${DRIVING_VIDEO_ABS%.*}"
TEMP_DRIVING="temp_driving_${TIMESTAMP}.${DRIVE_EXT}"
TEMP_DRIVING_PATH="${SCRIPT_DIR}LivePortrait-main/${TEMP_DRIVING}"

echo "Creating temporary copy of driving video: ${TEMP_DRIVING}"
echo "From: ${DRIVING_VIDEO_ABS}"
echo "To: ${TEMP_DRIVING_PATH}"

# 复制文件（对应bat的copy命令）
cp "${DRIVING_VIDEO_ABS}" "${TEMP_DRIVING_PATH}"

if [ $? -ne 0 ]; then
    echo "Error: Failed to copy driving video"
    echo "Check if file exists: ${DRIVING_VIDEO_ABS}"
    cd "${SCRIPT_DIR}" || exit 1
    exit 1
fi

# 进入LivePortrait目录
cd "${SCRIPT_DIR}LivePortrait-main" || exit 1

# 处理源图片路径（移除LivePortrait-main前缀）
LP_SOURCE="${SOURCE_IMAGE}"
LP_SOURCE=${LP_SOURCE//LivePortrait-main\//}
LP_SOURCE=${LP_SOURCE//LivePortrait-main\\/}

echo "Running LivePortrait with:"
echo "  Source: ${LP_SOURCE}"
echo "  Driving: ${TEMP_DRIVING}"
echo

# 运行 LivePortrait 推理（确保使用 live conda 环境）
source "$(conda info --base)"/etc/profile.d/conda.sh
conda activate live
python inference.py -s "${LP_SOURCE}" -d "${TEMP_DRIVING}"

if [ $? -ne 0 ]; then
    echo "Error: LivePortrait failed!"
    # 清理临时文件
    if [ -f "${TEMP_DRIVING_PATH}" ]; then
        rm "${TEMP_DRIVING_PATH}"
    fi
    cd "${SCRIPT_DIR}" || exit 1
    exit 1
fi

# 清理临时驱动视频
if [ -f "${TEMP_DRIVING_PATH}" ]; then
    rm "${TEMP_DRIVING_PATH}"
fi
echo "Temporary file deleted."

# 查找生成的视频文件
# 获取源图片的basename（支持.jpg 和.png 等扩展名）
SOURCE_BASENAME=$(basename "${SOURCE_IMAGE}" | sed 's/\.[^.]*$//')
EXPECTED_VIDEO="${SOURCE_BASENAME}--temp_driving_${TIMESTAMP}.mp4"
GENERATED_VIDEO=""

# 检查预期视频是否存在
if [ -f "${SCRIPT_DIR}LivePortrait-main/animations/${EXPECTED_VIDEO}" ]; then
    GENERATED_VIDEO="${SCRIPT_DIR}LivePortrait-main/animations/${EXPECTED_VIDEO}"
    echo "Found expected face-only video: ${EXPECTED_VIDEO}"
elif [ -n "$(ls "${SCRIPT_DIR}LivePortrait-main/animations/"*--temp_driving_${TIMESTAMP}.mp4 2>/dev/null | grep -v "_concat" | head -1)" ]; then
    # 查找最新生成的视频（不含_concat）
    GENERATED_VIDEO=$(ls -t "${SCRIPT_DIR}LivePortrait-main/animations/"*--temp_driving_${TIMESTAMP}.mp4 2>/dev/null | grep -v "_concat" | head -1)
    echo "Found video: $(basename "${GENERATED_VIDEO}")"
else
    # 降级查找：匹配*--output.mp4且不含_concat的文件
    FALLBACK_VIDEO=$(ls "${SCRIPT_DIR}LivePortrait-main/animations/"*--output.mp4 2>/dev/null | grep -v "_concat" | head -1)
    if [ -n "${FALLBACK_VIDEO}" ]; then
        GENERATED_VIDEO="${FALLBACK_VIDEO}"
        echo "Found fallback face-only video: $(basename "${GENERATED_VIDEO}")"
    else
        echo "Error: Could not find generated video in animations/"
        echo "Expected: ${EXPECTED_VIDEO}"
        echo "Searching in animations/..."
        ls "${SCRIPT_DIR}LivePortrait-main/animations/" 2>/dev/null
        cd "${SCRIPT_DIR}" || exit 1
        exit 1
    fi
fi

echo
echo "LivePortrait completed!"
echo "Generated Video: ${GENERATED_VIDEO}"
echo

cd "${SCRIPT_DIR}" || exit 1

# ========================================
# Step 2: Run genavatar.py for pre-processing
# ========================================
echo
echo "[Step 2/3] Running genavatar.py for pre-processing..."
echo

# 进入wav2lip目录
cd "${SCRIPT_DIR}wav2lip" || exit 1

echo "Generating avatar data with:"
echo "  Video: ${GENERATED_VIDEO}"
echo "  Avatar ID: ${AVATAR_ID}"
echo "  Image Size: 256"
echo

# 运行 genavatar.py
python genavatar.py --video_path "${GENERATED_VIDEO}" --img_size 256 --avatar_id "${AVATAR_ID}"

if [ $? -ne 0 ]; then
    echo "Error: genavatar.py failed!"
    cd "${SCRIPT_DIR}" || exit 1
    exit 1
fi

# 清理 LivePortrait-main 目录中的临时文件 (pkl 和视频)
echo "Cleaning up temporary files in LivePortrait-main..."
cd "${SCRIPT_DIR}LivePortrait-main" || exit 1

# 删除 temp_driving_*.pkl 文件
find . -maxdepth 1 -name "temp_driving_*.pkl" -type f -delete
echo "Temporary pkl files deleted."

# 删除 animations 目录中对应的视频文件
if [ -n "${GENERATED_VIDEO}" ] && [ -f "${GENERATED_VIDEO}" ]; then
    echo "Deleting generated video: ${GENERATED_VIDEO}"
    rm -f "${GENERATED_VIDEO}"
    
    # 同时删除 _concat 版本 (如果存在)
    CONCAT_VIDEO="${GENERATED_VIDEO%.mp4}_concat.mp4"
    if [ -f "${CONCAT_VIDEO}" ]; then
        echo "Deleting concat version: $(basename "${CONCAT_VIDEO}")"
        rm -f "${CONCAT_VIDEO}"
    fi
fi
echo "Temporary video files deleted."

cd "${SCRIPT_DIR}" || exit 1

echo
echo "Pre-processing completed!"
echo "Avatar data saved to: data/avatars/${AVATAR_ID}"
echo

# ========================================
# Step 3: Run app.py with edgetts for digital human generation
# ========================================
echo
echo "[Step 3/3] Starting app.py with edgetts for digital human generation..."
echo

echo "Starting digital human server..."
echo "Avatar ID: ${AVATAR_ID}"
echo "TTS: edgetts (Edge TTS via local API http://127.0.0.1:1024/tts)"
echo
echo "The server will start and you can access it via web interface."
echo "Press Ctrl+C to stop the server."
echo

# 检查并安装pyaudio（对应bat的pyaudio检查）
echo "Checking pyaudio installation..."
python -c "import pyaudio" 2>/dev/null || (echo "Installing pyaudio..." && pip install pyaudio)

# 启动 app.py
# 关键环境变量说明：
#   AIOICE_PORT_MIN/MAX  : aiortc 绑定的 UDP 端口范围，必须与防火墙放行范围一致
#   AIOICE_BIND_IP       : aiortc 只绑定这一个 IP（服务器有多网卡时，指定浏览器能路由到的那个）
#   WEBRTC_NAT_IP        : 如果浏览器用的是跳板机 IP 而非服务器 IP，设置此项做 SDP IP 替换
#
# 默认值（可通过外部传入环境变量覆盖）：
#   CUDA_VISIBLE_DEVICES=6,7  (使用 GPU 6,7 运行)
#   AIOICE_PORT_MIN=50000  AIOICE_PORT_MAX=50010
#   AIOICE_BIND_IP=192.168.22.9  WEBRTC_NAT_IP=183.252.196.135
#
echo "Using env: CUDA_VISIBLE_DEVICES=${CUDA_VISIBLE_DEVICES} AIOICE_BIND_IP=${AIOICE_BIND_IP} WEBRTC_NAT_IP=${WEBRTC_NAT_IP} AIOICE_PORT_MIN=${AIOICE_PORT_MIN} AIOICE_PORT_MAX=${AIOICE_PORT_MAX}"

CUDA_VISIBLE_DEVICES="${CUDA_VISIBLE_DEVICES}" \
AIOICE_BIND_IP="${AIOICE_BIND_IP}" \
WEBRTC_NAT_IP="${WEBRTC_NAT_IP}" \
AIOICE_PORT_MIN="${AIOICE_PORT_MIN}" \
AIOICE_PORT_MAX="${AIOICE_PORT_MAX}" \
python app.py \
    --avatar_id "${AVATAR_ID}" \
    --tts edgetts \
    --REF_FILE zh-CN-YunxiaNeural \
    --model wav2lip \
    --transport webrtc \
    --listenport 7868

echo
echo "========================================"
echo "Pipeline completed!"
echo "========================================"