bcut-asr


Namebcut-asr JSON
Version 0.0.3 PyPI version JSON
download
home_pageNone
Summary使用必剪API的语音字幕识别
upload_time2024-04-28 03:06:44
maintainerNone
docs_urlNone
authorSocialSisterYi
requires_python>=3.10
licenseMIT
keywords
VCS
bugtrack_url
requirements No requirements were recorded.
Travis-CI No Travis.
coveralls test coverage No coveralls. 
            <h1 align="center">Bcut-ASR</h1>

使用必剪 API 进行云端语音字幕识别，支持 CLI 和 module 调用

## ✨Feature

- 可直接上传`flac`, `aac`, `m4a`, `mp3`, `wav`音频格式
- 自动调用 ffmpeg, 实现视频伴音和其他音频格式转码
- 支持`srt`, `json`, `lrc`, `txt`格式字幕输出
- 字幕支持断句和首位时间标记
- 可使用 stdout 输出字幕文本

## 🚀Install

首先确保 ffmpeg 已安装，且 PATH 中可以访问，若未安装可以使用如下命令（已安装请无视）：

Linux：

```bash
sudo apt install ffmpeg
```

Windows：

```powershell
winget install ffmpeg
```

本项目暂时未发布 pypi，应使用本地安装，Python 版本应 >= 3.10，需要安装 poetry 

```bash
git clone https://github.com/SocialSisterYi/bcut-asr
cd bcut-asr
poetry build -f wheel
pip install dist/bcut_asr-0.0.3-py3-none-any.whl # Example
```

## 📃Usage

### CLI Interface

```bash
bcut_asr video.mp4
```

或

```bash
bcut_asr video.mp4 subtitle.srt
```

或

```bash
bcut_asr video.mp4 -f srt - > subtitle.srt
```

长音频指定任务状态轮询间隔(秒)，避免接口频繁调用

```bash
bcut_asr video.mp4 -f srt -i 30 - > subtitle.srt
```

```
bcut_asr -h
usage: bcut-asr [-h] [-f [{srt,json,lrc,txt}]] [-i [1.0]] input [output]

必剪语音识别

positional arguments:
  input                 输入媒体文件
  output                输出字幕文件, 可stdout

options:
  -h, --help            show this help message and exit
  -f [{srt,json,lrc,txt}], --format [{srt,json,lrc,txt}]
                        输出字幕格式
  -i [1.0], --interval [1.0]
                        任务状态轮询间隔(秒)

支持输入音频格式: flac, aac, m4a, mp3, wav 支持自动调用ffmpeg提取视频伴音
```

### Module

```python
from bcut_asr import BcutASR
from bcut_asr.orm import ResultStateEnum

asr = BcutASR('voice.mp3')
asr.upload() # 上传文件
asr.create_task() # 创建任务

# 轮询检查结果
while True:
    result = asr.result()
    # 判断识别成功
    if result.state == ResultStateEnum.COMPLETE:
        break

# 解析字幕内容
subtitle = result.parse()
# 判断是否存在字幕
if subtitle.has_data():
    # 输出srt格式
    print(subtitle.to_srt())
```

输入视频

```python
from bcut_asr import run_everywhere
from argparse import Namespace


f = open("file.mp4", "rb")
argg = Namespace(format="srt", interval=30.0, input=f, output=None)
run_everywhere(argg)

```

Raw data

            {
    "_id": null,
    "home_page": null,
    "name": "bcut-asr",
    "maintainer": null,
    "docs_url": null,
    "requires_python": ">=3.10",
    "maintainer_email": null,
    "keywords": null,
    "author": "SocialSisterYi",
    "author_email": "1440239038@qq.com",
    "download_url": null,
    "platform": null,
    "description": "<h1 align=\"center\">Bcut-ASR</h1>\n\n\u4f7f\u7528\u5fc5\u526a API \u8fdb\u884c\u4e91\u7aef\u8bed\u97f3\u5b57\u5e55\u8bc6\u522b\uff0c\u652f\u6301 CLI \u548c module \u8c03\u7528\n\n## \u2728Feature\n\n- \u53ef\u76f4\u63a5\u4e0a\u4f20`flac`, `aac`, `m4a`, `mp3`, `wav`\u97f3\u9891\u683c\u5f0f\n- \u81ea\u52a8\u8c03\u7528 ffmpeg, \u5b9e\u73b0\u89c6\u9891\u4f34\u97f3\u548c\u5176\u4ed6\u97f3\u9891\u683c\u5f0f\u8f6c\u7801\n- \u652f\u6301`srt`, `json`, `lrc`, `txt`\u683c\u5f0f\u5b57\u5e55\u8f93\u51fa\n- \u5b57\u5e55\u652f\u6301\u65ad\u53e5\u548c\u9996\u4f4d\u65f6\u95f4\u6807\u8bb0\n- \u53ef\u4f7f\u7528 stdout \u8f93\u51fa\u5b57\u5e55\u6587\u672c\n\n## \ud83d\ude80Install\n\n\u9996\u5148\u786e\u4fdd ffmpeg \u5df2\u5b89\u88c5\uff0c\u4e14 PATH \u4e2d\u53ef\u4ee5\u8bbf\u95ee\uff0c\u82e5\u672a\u5b89\u88c5\u53ef\u4ee5\u4f7f\u7528\u5982\u4e0b\u547d\u4ee4\uff08\u5df2\u5b89\u88c5\u8bf7\u65e0\u89c6\uff09\uff1a\n\nLinux\uff1a\n\n```bash\nsudo apt install ffmpeg\n```\n\nWindows\uff1a\n\n```powershell\nwinget install ffmpeg\n```\n\n\u672c\u9879\u76ee\u6682\u65f6\u672a\u53d1\u5e03 pypi\uff0c\u5e94\u4f7f\u7528\u672c\u5730\u5b89\u88c5\uff0cPython \u7248\u672c\u5e94 >= 3.10\uff0c\u9700\u8981\u5b89\u88c5 poetry \n\n```bash\ngit clone https://github.com/SocialSisterYi/bcut-asr\ncd bcut-asr\npoetry build -f wheel\npip install dist/bcut_asr-0.0.3-py3-none-any.whl # Example\n```\n\n## \ud83d\udcc3Usage\n\n### CLI Interface\n\n```bash\nbcut_asr video.mp4\n```\n\n\u6216\n\n```bash\nbcut_asr video.mp4 subtitle.srt\n```\n\n\u6216\n\n```bash\nbcut_asr video.mp4 -f srt - > subtitle.srt\n```\n\n\u957f\u97f3\u9891\u6307\u5b9a\u4efb\u52a1\u72b6\u6001\u8f6e\u8be2\u95f4\u9694(\u79d2)\uff0c\u907f\u514d\u63a5\u53e3\u9891\u7e41\u8c03\u7528\n\n```bash\nbcut_asr video.mp4 -f srt -i 30 - > subtitle.srt\n```\n\n```\nbcut_asr -h\nusage: bcut-asr [-h] [-f [{srt,json,lrc,txt}]] [-i [1.0]] input [output]\n\n\u5fc5\u526a\u8bed\u97f3\u8bc6\u522b\n\npositional arguments:\n  input                 \u8f93\u5165\u5a92\u4f53\u6587\u4ef6\n  output                \u8f93\u51fa\u5b57\u5e55\u6587\u4ef6, \u53efstdout\n\noptions:\n  -h, --help            show this help message and exit\n  -f [{srt,json,lrc,txt}], --format [{srt,json,lrc,txt}]\n                        \u8f93\u51fa\u5b57\u5e55\u683c\u5f0f\n  -i [1.0], --interval [1.0]\n                        \u4efb\u52a1\u72b6\u6001\u8f6e\u8be2\u95f4\u9694(\u79d2)\n\n\u652f\u6301\u8f93\u5165\u97f3\u9891\u683c\u5f0f: flac, aac, m4a, mp3, wav \u652f\u6301\u81ea\u52a8\u8c03\u7528ffmpeg\u63d0\u53d6\u89c6\u9891\u4f34\u97f3\n```\n\n### Module\n\n```python\nfrom bcut_asr import BcutASR\nfrom bcut_asr.orm import ResultStateEnum\n\nasr = BcutASR('voice.mp3')\nasr.upload() # \u4e0a\u4f20\u6587\u4ef6\nasr.create_task() # \u521b\u5efa\u4efb\u52a1\n\n# \u8f6e\u8be2\u68c0\u67e5\u7ed3\u679c\nwhile True:\n    result = asr.result()\n    # \u5224\u65ad\u8bc6\u522b\u6210\u529f\n    if result.state == ResultStateEnum.COMPLETE:\n        break\n\n# \u89e3\u6790\u5b57\u5e55\u5185\u5bb9\nsubtitle = result.parse()\n# \u5224\u65ad\u662f\u5426\u5b58\u5728\u5b57\u5e55\nif subtitle.has_data():\n    # \u8f93\u51fasrt\u683c\u5f0f\n    print(subtitle.to_srt())\n```\n\n\u8f93\u5165\u89c6\u9891\n\n```python\nfrom bcut_asr import run_everywhere\nfrom argparse import Namespace\n\n\nf = open(\"file.mp4\", \"rb\")\nargg = Namespace(format=\"srt\", interval=30.0, input=f, output=None)\nrun_everywhere(argg)\n\n```\n",
    "bugtrack_url": null,
    "license": "MIT",
    "summary": "\u4f7f\u7528\u5fc5\u526aAPI\u7684\u8bed\u97f3\u5b57\u5e55\u8bc6\u522b",
    "version": "0.0.3",
    "project_urls": null,
    "split_keywords": [],
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "blake2b_256": "1729e4bf1bef9dd5c56e759df97eb55c2f57526f70b577c37873da021b84dbb6",
                "md5": "9e16f62d7d2807230166e585c7ebb5d0",
                "sha256": "5f74e7d3c5f05c96d1e71d82d93c34388c8f2c797f3cf03f0a392f269e5f729f"
            },
            "downloads": -1,
            "filename": "bcut_asr-0.0.3-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "9e16f62d7d2807230166e585c7ebb5d0",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": ">=3.10",
            "size": 9770,
            "upload_time": "2024-04-28T03:06:44",
            "upload_time_iso_8601": "2024-04-28T03:06:44.837619Z",
            "url": "https://files.pythonhosted.org/packages/17/29/e4bf1bef9dd5c56e759df97eb55c2f57526f70b577c37873da021b84dbb6/bcut_asr-0.0.3-py3-none-any.whl",
            "yanked": false,
            "yanked_reason": null
        }
    ],
    "upload_time": "2024-04-28 03:06:44",
    "github": false,
    "gitlab": false,
    "bitbucket": false,
    "codeberg": false,
    "lcname": "bcut-asr"
}
SocialSisterYi
Elapsed time: 1.75119s