feat: add automatic audio extraction for video downloads

base/base_crawler.py

@@ -116,6 +116,12 @@ async def store_video(self, video_content_item: Dict):
         pass
 
 
+class AbstractStoreAudio(ABC):
+    # TODO: support all platform
+    async def store_audio(self, audio_content_item: Dict):
+        pass
+
+
 class AbstractApiClient(ABC):
 
     @abstractmethod

config/base_config.py

@@ -107,6 +107,17 @@
 # Whether to enable crawling media mode (including image or video resources), crawling media is not enabled by default
 ENABLE_GET_MEIDAS = False
 
+# Whether to enable audio download/extraction mode.
+# When True, the crawler will automatically extract audio from downloaded videos and save it as a separate file.
+ENABLE_GET_AUDIO = False
+
+# Audio output format, supports: mp3, aac
+AUDIO_FORMAT = "mp3"
+
+# Whether to keep the original video file after extracting audio.
+# Set to False to delete the video and keep only the audio, saving disk space.
+KEEP_ORIGINAL_VIDEO = True
+
 # Whether to enable comment crawling mode. Comment crawling is enabled by default.
 ENABLE_GET_COMMENTS = True
 

media_platform/bilibili/core.py

@@ -607,6 +607,11 @@ async def get_bilibili_video(self, video_item: Dict, semaphore: asyncio.Semaphor
         extension_file_name = f"video.mp4"
         await bilibili_store.store_video(aid, content, extension_file_name)
 
+        if config.ENABLE_GET_AUDIO:
+            video_path = f"{bilibili_store.BilibiliVideo().video_store_path}/{aid}/{extension_file_name}"
+            audio_extension_file_name = f"{extension_file_name.rsplit('.', 1)[0]}.{config.AUDIO_FORMAT}"
+            await bilibili_store.store_audio(aid, video_path, audio_extension_file_name)
+
     async def get_all_creator_details(self, creator_url_list: List[str]):
         """
         creator_url_list: get details for creator from creator URL list

media_platform/douyin/core.py

@@ -468,3 +468,8 @@ async def get_aweme_video(self, aweme_item: Dict):
             return
         extension_file_name = f"video.mp4"
         await douyin_store.update_dy_aweme_video(aweme_id, content, extension_file_name)
+
+        if config.ENABLE_GET_AUDIO:
+            video_path = f"{douyin_store.DouYinVideo().video_store_path}/{aweme_id}/{extension_file_name}"
+            audio_extension_file_name = f"{extension_file_name.rsplit('.', 1)[0]}.{config.AUDIO_FORMAT}"
+            await douyin_store.update_dy_aweme_audio(aweme_id, video_path, audio_extension_file_name)

media_platform/xhs/core.py

@@ -520,3 +520,8 @@ async def get_notice_video(self, note_item: Dict):
             extension_file_name = f"{videoNum}.mp4"
             videoNum += 1
             await xhs_store.update_xhs_note_video(note_id, content, extension_file_name)
+
+            if config.ENABLE_GET_AUDIO:
+                video_path = f"{xhs_store.XiaoHongShuVideo().video_store_path}/{note_id}/{extension_file_name}"
+                audio_extension_file_name = f"{extension_file_name.rsplit('.', 1)[0]}.{config.AUDIO_FORMAT}"
+                await xhs_store.update_xhs_note_audio(note_id, video_path, audio_extension_file_name)

store/bilibili/__init__.py

@@ -30,6 +30,7 @@
 
 from ._store_impl import *
 from .bilibilli_store_media import *
+from .bilibili_store_audio import *
 
 
 class BiliStoreFactory:
@@ -131,6 +132,24 @@ async def store_video(aid, video_content, extension_file_name):
     })
 
 
+async def store_audio(aid, video_path, extension_file_name):
+    """
+    Extract and save Bilibili video audio from a downloaded video.
+    Args:
+        aid:
+        video_path:
+        extension_file_name:
+
+    Returns:
+        Path to the extracted audio file, or None if failed.
+    """
+    return await BilibiliAudio().store_audio({
+        "aid": aid,
+        "video_path": video_path,
+        "extension_file_name": extension_file_name,
+    })
+
+
 async def batch_update_bilibili_creator_fans(creator_info: Dict, fans_list: List[Dict]):
     # 教学版：不再采集/存储粉丝列表(其他用户的个人信息)，防骚扰。
     return

store/bilibili/bilibili_store_audio.py

@@ -0,0 +1,76 @@
+# -*- coding: utf-8 -*-
+# Copyright (c) 2025 relakkes@gmail.com
+#
+# This file is part of MediaCrawler project.
+# Repository: https://github.com/NanmiCoder/MediaCrawler/blob/main/store/bilibili/bilibili_store_audio.py
+# GitHub: https://github.com/NanmiCoder
+# Licensed under NON-COMMERCIAL LEARNING LICENSE 1.1
+#
+# 声明：本代码仅供学习和研究目的使用。使用者应遵守以下原则：
+# 1. 不得用于任何商业用途。
+# 2. 使用时应遵守目标平台的使用条款和robots.txt规则。
+# 3. 不得进行大规模爬取或对目标平台造成运营干扰。
+# 4. 应合理控制请求频率，避免给目标平台带来不必要的负担。
+# 5. 不得用于任何非法或不当的用途。
+#
+# 详细许可条款请参阅项目根目录下的LICENSE文件。
+# 使用本代码即表示您同意遵守上述原则和LICENSE中的所有条款。
+
+from pathlib import Path
+from typing import Dict, Optional
+
+from base.base_crawler import AbstractStoreAudio
+from tools import utils
+from tools.media_util import extract_audio_from_video
+import config
+
+
+class BilibiliAudio(AbstractStoreAudio):
+    def __init__(self):
+        if config.SAVE_DATA_PATH:
+            self.audio_store_path = f"{config.SAVE_DATA_PATH}/bili/audios"
+        else:
+            self.audio_store_path = "data/bili/audios"
+
+    def make_save_file_name(self, aid: str, extension_file_name: str) -> str:
+        """
+        make save file name by store type
+
+        Args:
+            aid: aid
+            extension_file_name: audio filename with extension
+
+        Returns:
+
+        """
+        return f"{self.audio_store_path}/{aid}/{extension_file_name}"
+
+    async def store_audio(self, audio_content_item: Dict) -> Optional[str]:
+        """
+        Extract audio from a downloaded video and save it to the audio store path.
+
+        Args:
+            audio_content_item: dict with keys:
+                - aid: aid
+                - video_path: path to the downloaded video file
+                - extension_file_name: audio filename with extension
+
+        Returns:
+            Path to the extracted audio file, or None if failed.
+        """
+        aid: str = str(audio_content_item.get("aid"))
+        video_path: str = audio_content_item.get("video_path")
+        extension_file_name: str = audio_content_item.get("extension_file_name")
+
+        if not video_path or not Path(video_path).exists():
+            utils.logger.warning(f"[BilibiliAudio.store_audio] Video not found: {video_path}")
+            return None
+
+        Path(self.audio_store_path + "/" + aid).mkdir(parents=True, exist_ok=True)
+        audio_path = self.make_save_file_name(aid, extension_file_name)
+
+        return await extract_audio_from_video(
+            video_path=video_path,
+            audio_path=audio_path,
+            audio_format=config.AUDIO_FORMAT,
+        )

store/douyin/__init__.py

@@ -29,6 +29,7 @@
 
 from ._store_impl import *
 from .douyin_store_media import *
+from .douyin_store_audio import *
 
 
 class DouyinStoreFactory:
@@ -249,3 +250,18 @@ async def update_dy_aweme_video(aweme_id, video_content, extension_file_name):
     """
 
     await DouYinVideo().store_video({"aweme_id": aweme_id, "video_content": video_content, "extension_file_name": extension_file_name})
+
+
+async def update_dy_aweme_audio(aweme_id, video_path, extension_file_name):
+    """
+    Extract and save Douyin aweme audio from a downloaded video.
+    Args:
+        aweme_id:
+        video_path:
+        extension_file_name:
+
+    Returns:
+        Path to the extracted audio file, or None if failed.
+    """
+
+    return await DouYinAudio().store_audio({"aweme_id": aweme_id, "video_path": video_path, "extension_file_name": extension_file_name})

store/douyin/douyin_store_audio.py

@@ -0,0 +1,76 @@
+# -*- coding: utf-8 -*-
+# Copyright (c) 2025 relakkes@gmail.com
+#
+# This file is part of MediaCrawler project.
+# Repository: https://github.com/NanmiCoder/MediaCrawler/blob/main/store/douyin/douyin_store_audio.py
+# GitHub: https://github.com/NanmiCoder
+# Licensed under NON-COMMERCIAL LEARNING LICENSE 1.1
+#
+# 声明：本代码仅供学习和研究目的使用。使用者应遵守以下原则：
+# 1. 不得用于任何商业用途。
+# 2. 使用时应遵守目标平台的使用条款和robots.txt规则。
+# 3. 不得进行大规模爬取或对目标平台造成运营干扰。
+# 4. 应合理控制请求频率，避免给目标平台带来不必要的负担。
+# 5. 不得用于任何非法或不当的用途。
+#
+# 详细许可条款请参阅项目根目录下的LICENSE文件。
+# 使用本代码即表示您同意遵守上述原则和LICENSE中的所有条款。
+
+from pathlib import Path
+from typing import Dict, Optional
+
+from base.base_crawler import AbstractStoreAudio
+from tools import utils
+from tools.media_util import extract_audio_from_video
+import config
+
+
+class DouYinAudio(AbstractStoreAudio):
+    def __init__(self):
+        if config.SAVE_DATA_PATH:
+            self.audio_store_path = f"{config.SAVE_DATA_PATH}/douyin/audios"
+        else:
+            self.audio_store_path = "data/douyin/audios"
+
+    def make_save_file_name(self, aweme_id: str, extension_file_name: str) -> str:
+        """
+        make save file name by store type
+
+        Args:
+            aweme_id: aweme id
+            extension_file_name: audio filename with extension
+
+        Returns:
+
+        """
+        return f"{self.audio_store_path}/{aweme_id}/{extension_file_name}"
+
+    async def store_audio(self, audio_content_item: Dict) -> Optional[str]:
+        """
+        Extract audio from a downloaded video and save it to the audio store path.
+
+        Args:
+            audio_content_item: dict with keys:
+                - aweme_id: aweme id
+                - video_path: path to the downloaded video file
+                - extension_file_name: audio filename with extension
+
+        Returns:
+            Path to the extracted audio file, or None if failed.
+        """
+        aweme_id: str = audio_content_item.get("aweme_id")
+        video_path: str = audio_content_item.get("video_path")
+        extension_file_name: str = audio_content_item.get("extension_file_name")
+
+        if not video_path or not Path(video_path).exists():
+            utils.logger.warning(f"[DouYinAudio.store_audio] Video not found: {video_path}")
+            return None
+
+        Path(self.audio_store_path + "/" + aweme_id).mkdir(parents=True, exist_ok=True)
+        audio_path = self.make_save_file_name(aweme_id, extension_file_name)
+
+        return await extract_audio_from_video(
+            video_path=video_path,
+            audio_path=audio_path,
+            audio_format=config.AUDIO_FORMAT,
+        )

store/xhs/__init__.py

@@ -28,6 +28,7 @@
 from tools.user_hash import anonymize_user_id, mask_nickname
 
 from .xhs_store_media import *
+from .xhs_store_audio import *
 from ._store_impl import *
 
 
@@ -220,3 +221,18 @@ async def update_xhs_note_video(note_id, video_content, extension_file_name):
     """
 
     await XiaoHongShuVideo().store_video({"notice_id": note_id, "video_content": video_content, "extension_file_name": extension_file_name})
+
+
+async def update_xhs_note_audio(note_id, video_path, extension_file_name):
+    """
+    Extract and save Xiaohongshu note audio from a downloaded video.
+    Args:
+        note_id:
+        video_path:
+        extension_file_name:
+
+    Returns:
+        Path to the extracted audio file, or None if failed.
+    """
+
+    return await XiaoHongShuAudio().store_audio({"notice_id": note_id, "video_path": video_path, "extension_file_name": extension_file_name})

store/xhs/xhs_store_audio.py

@@ -0,0 +1,76 @@
+# -*- coding: utf-8 -*-
+# Copyright (c) 2025 relakkes@gmail.com
+#
+# This file is part of MediaCrawler project.
+# Repository: https://github.com/NanmiCoder/MediaCrawler/blob/main/store/xhs/xhs_store_audio.py
+# GitHub: https://github.com/NanmiCoder
+# Licensed under NON-COMMERCIAL LEARNING LICENSE 1.1
+#
+# 声明：本代码仅供学习和研究目的使用。使用者应遵守以下原则：
+# 1. 不得用于任何商业用途。
+# 2. 使用时应遵守目标平台的使用条款和robots.txt规则。
+# 3. 不得进行大规模爬取或对目标平台造成运营干扰。
+# 4. 应合理控制请求频率，避免给目标平台带来不必要的负担。
+# 5. 不得用于任何非法或不当的用途。
+#
+# 详细许可条款请参阅项目根目录下的LICENSE文件。
+# 使用本代码即表示您同意遵守上述原则和LICENSE中的所有条款。
+
+from pathlib import Path
+from typing import Dict, Optional
+
+from base.base_crawler import AbstractStoreAudio
+from tools import utils
+from tools.media_util import extract_audio_from_video
+import config
+
+
+class XiaoHongShuAudio(AbstractStoreAudio):
+    def __init__(self):
+        if config.SAVE_DATA_PATH:
+            self.audio_store_path = f"{config.SAVE_DATA_PATH}/xhs/audios"
+        else:
+            self.audio_store_path = "data/xhs/audios"
+
+    def make_save_file_name(self, notice_id: str, extension_file_name: str) -> str:
+        """
+        make save file name by store type
+
+        Args:
+            notice_id: notice id
+            extension_file_name: audio filename with extension
+
+        Returns:
+
+        """
+        return f"{self.audio_store_path}/{notice_id}/{extension_file_name}"
+
+    async def store_audio(self, audio_content_item: Dict) -> Optional[str]:
+        """
+        Extract audio from a downloaded video and save it to the audio store path.
+
+        Args:
+            audio_content_item: dict with keys:
+                - notice_id: note id
+                - video_path: path to the downloaded video file
+                - extension_file_name: audio filename with extension
+
+        Returns:
+            Path to the extracted audio file, or None if failed.
+        """
+        notice_id: str = audio_content_item.get("notice_id")
+        video_path: str = audio_content_item.get("video_path")
+        extension_file_name: str = audio_content_item.get("extension_file_name")
+
+        if not video_path or not Path(video_path).exists():
+            utils.logger.warning(f"[XiaoHongShuAudio.store_audio] Video not found: {video_path}")
+            return None
+
+        Path(self.audio_store_path + "/" + notice_id).mkdir(parents=True, exist_ok=True)
+        audio_path = self.make_save_file_name(notice_id, extension_file_name)
+
+        return await extract_audio_from_video(
+            video_path=video_path,
+            audio_path=audio_path,
+            audio_format=config.AUDIO_FORMAT,
+        )