osc-data

多模态数据表示与处理库，提供 Python 友好的 API 和 Rust 实现的高性能核心。

简介

osc-data 是一个用于多模态数据处理（文本、图像、音频、视频）的 Python 库，核心计算模块使用 Rust 实现以提供高性能。

特性

文本处理：文本正则化（TN）、去 emoji、全角转半角、流式句子分割，支持中英文混合文本
图像处理：加载、保存、格式转换、缩放、裁剪，支持 PNG/JPEG/WebP/BMP/GIF/TIFF
音频处理：音频加载、特征提取（分贝计算等）
视频处理：视频加载、帧提取、关键帧分割、音频提取与合并、保存；支持 MP4/MOV 等 FFmpeg 可解封装格式
音视频同步：音频时长自动调整（循环/静音填充）、音频截断警告
高性能：核心算法使用 Rust + PyO3 实现，Python 端使用 DocArray 类型系统

安装

环境要求

Python >= 3.9
Rust >= 1.70 (编译时需要，仅源码安装需要)

使用 pip 安装

# 从 PyPI 安装 
pip install osc-data --upgrade

# 从 Git 仓库直接安装
pip install git+https://github.com/di-osc/osc-data.git

# 安装指定版本（示例）
pip install "osc-data>=0.3.2"

从源码安装

# 克隆仓库
git clone https://github.com/di-osc/osc-data.git
cd osc-data

# 使用 uv 安装 (推荐)
uv sync

# 或使用 pip + maturin 构建
pip install maturin
maturin develop --release

# 或以可编辑模式安装
pip install -e .

快速开始

图像处理

from osc_data.image import Image

# 从本地或 URL 加载图片
img = Image(uri="./photo.jpg").load()
# img = Image(uri="https://example.com/image.png").load()

# 查看图片信息
print(f"尺寸: {img.width}x{img.height}, 色彩模式: {img.color_mode}")
print(f"源格式: {img.source_format}")  # png, jpeg 等

# 格式转换
rgb_img = img.to_rgb()

# 缩放 (使用 Lanczos3 高质量算法)
small_img = img.resize(256, 256)

# 裁剪
cropped = img.crop(100, 100, 200, 200)

# 保存 (使用 file_format 指定输出格式)
img.save("output.png")
img.save("output.jpg", file_format="jpeg", quality=95)

# 批量处理（传入已加载的 Image 实例列表）
# images = [img_a, img_b, img_c]
# resized_batch = Image.batch_resize_images(images, 256, 256)

文本处理

from osc_data.text import TextNormalizer, TextCleaner

# 文本正则化（TN）：数字、日期、货币等转为口语化表达
normalizer = TextNormalizer()
result = normalizer.normalize("我有100元")
print(result)  # "我有一百元"

# 可选参数
normalizer = TextNormalizer(
    remove_erhua=True,   # 去除儿化音
    remove_emoji=True,   # 去除 emoji
    to_half_width=True,  # 全角转半角
)

# 轻量文本清洗（不做 TN）
cleaner = TextCleaner(remove_emoji=True, to_half_width=True)
result = cleaner.clean("Ｈｅｌｌｏ！😀你好")
print(result)  # "Hello!你好"

文本流处理

from osc_data.text_stream import TextStreamSentencizer

sentencizer = TextStreamSentencizer()

# 流式处理文本
streaming_text = "你好！这是第一句话。这是第二句话。"

sentences = []
for char in streaming_text:
    sentences.extend(sentencizer.push(char))

# 输出剩余内容
sentences.extend(sentencizer.flush())

print(sentences)
# ['你好！', '这是第一句话。', '这是第二句话。']

视频处理

from osc_data.video import Video
from osc_data.audio import Audio

# 加载视频（或使用示例视频）
video = Video(uri="./video.mp4").load()
# video = Video().load_example()  # 加载内置示例视频

print(f"分辨率: {video.width}x{video.height}, FPS: {video.fps}, 时长: {video.duration}s")

# 按关键帧分割
segments = video.split_by_key_frames(min_split_duration_s=5)

# 提取音频
audio = video.extract_audio()

# 合并音频（自动调整音频时长匹配视频）
# 指定 output_path 时会把已含音视频的完整文件复制到目标路径（保留音轨）；
# 勿用仅含帧数据的 merged.save() 代替导出，否则输出的文件没有音轨。
new_audio = Audio(uri="./music.mp3").load()
merged = video.merge_audio(new_audio, "output.mp4", audio_mode="loop")
# audio_mode: "loop" 循环填充, "silence" 静音填充

# 移除音频
no_audio = video.remove_audio("silent.mp4")

# 静态方法合并视频和音频
combined = Video.combine_video_audio(video, new_audio, "final.mp4")

# 在 Jupyter 中显示视频
video.display()

# 根据宽高比计算最佳尺寸
best_w, best_h = video.get_best_size((9, 16))
print(f"9:16 最佳尺寸: {best_w}x{best_h}")

# 保存视频（仅编码内存中的视频帧，不含音轨；目标路径已存在时会覆盖）
video.save("output.mp4", format="mp4", codec="h264")

音频处理

from osc_data.audio import Audio

# 加载音频
audio = Audio(uri="./audio.wav").load()

print(f"采样率: {audio.sample_rate} Hz, 时长: {audio.duration_s:.2f} s")

# 计算分贝
from osc_data._core import compute_decibel

db = compute_decibel(audio.data, audio.sample_rate)

API 文档

Image 类

属性/方法	说明
`uri`	图片路径或 URL
`color_mode`	色彩模式 (RGB/RGBA/L)
`source_format`	源文件格式 (png/jpeg/webp)
`width`, `height`	图片尺寸
`load()`	从本地路径或 URL 加载图片
`save(path, file_format, quality)`	保存图片到本地
`to_rgb()`	转换为 RGB 格式
`resize(width, height)`	缩放图片 (Lanczos3)
`crop(x, y, width, height)`	裁剪图片
`to_bytes(file_format, quality)`	转换为字节
`from_bytes(data)`	从字节创建图片
`batch_resize_images(images, width, height)`	批量缩放
`display()`	显示图片 (Jupyter)

TextNormalizer 类

属性/方法	说明
`remove_erhua`	是否去除儿化音，默认 False
`remove_emoji`	是否去除 emoji，默认 False
`to_half_width`	是否全角转半角，默认 False
`normalize(text)`	执行文本正则化（TN），返回口语化文本

TextCleaner 类

属性/方法	说明
`remove_emoji`	是否去除 emoji，默认 False
`to_half_width`	是否全角转半角，默认 False
`clean(text)`	执行文本清洗，返回清洗后文本

TextStreamSentencizer 类

方法	说明
`push(char)`	推入单个字符，返回已完成的句子列表
`flush()`	清空缓冲区，返回剩余内容
`reset()`	重置状态

Audio 类

属性/方法	说明
`uri`	音频路径或 URL
`sample_rate`	采样率（Hz）
`data`	`float32` 波形数组；单声道为 `(samples,)`，多声道为 `(channels, samples)`
`load(sample_rate, mono)`	加载；可指定目标采样率与是否单声道
`duration_s` / `duration_ms`	时长（秒 / 毫秒）
`save(path, format)`	经 PyAV 编码保存；支持 MP3、WAV、FLAC、OGG 等

Video 类

属性/方法	说明
`uri`	视频路径或 URL
`width`, `height`	视频分辨率
`fps`	帧率
`duration`	时长（秒）
`has_audio`	是否有音轨
`load()`	从本地路径或 URL 加载视频（仅解码视频帧至内存）
`load_example()`	加载内置示例视频
`save(path, format, codec)`	保存视频（仅视频帧；路径存在则覆盖）
`display()`	显示视频 (Jupyter)
`split_by_key_frames(min_split_duration_s)`	按关键帧分割
`extract_audio()`	提取音频
`merge_audio(audio, output_path, audio_mode)`	合并音频；写盘时复制完整封装以保留音轨
`get_best_size(ratio)`	根据宽高比计算最佳尺寸
`remove_audio(output_path)`	移除音频
`combine_video_audio(video, audio, output_path)`	静态方法：合并视频和音频

音频时长处理：

audio_mode="loop"：循环播放音频填充（默认）
audio_mode="silence"：静音填充
音频长于视频时自动截断并发出警告

项目结构

osc-data/
├── osc_data/           # Python 模块
│   ├── __init__.py
│   ├── image.py        # 图像处理
│   ├── video.py        # 视频处理
│   ├── audio.py        # 音频处理
│   ├── text.py         # 文本处理
│   ├── text_stream.py  # 流式文本分割
│   └── assets/         # 示例资源
│       ├── image/      # 示例图片
│       ├── audio/      # 示例音频
│       ├── video/      # 示例视频
│       └── text/       # 示例文本
├── src/                # Rust 核心模块
│   ├── lib.rs          # 模块入口
│   ├── image.rs        # 图像处理核心
│   ├── audio.rs        # 音频处理核心
│   ├── text.rs         # 文本处理核心
│   ├── text_stream.rs  # 流式文本分割核心
│   └── reorder.rs      # 重排序等辅助逻辑
├── tests/              # 测试文件
│   ├── test_image.py
│   ├── test_sentencizer.py
│   └── ...
├── Cargo.toml          # Rust 配置
├── pyproject.toml      # Python 配置
└── README.md           # 本文件

开发

构建 Rust 扩展

maturin develop        # 开发模式 (快速编译)
maturin develop --release  # 发布模式 (优化)

运行测试

# 运行所有测试
pytest

# 运行特定模块测试
pytest tests/test_image.py -v
pytest tests/test_video.py -v
pytest tests/test_sentencizer.py -v

代码风格

# Python 代码格式化
ruff format osc_data/ tests/

# Rust 代码格式化
cargo fmt

依赖

Python 依赖

kaldifst >= 1.7.14: FST / 文本相关依赖
pydantic >= 2.11.7: 数据验证
docarray: 多模态数据类型系统
numpy: 数组操作
requests: HTTP 请求
av >= 10.0.0: 视频/音频编解码
wasabi >= 1.1.0: 格式化日志输出

Rust 依赖

pyo3 >= 0.25.0: Python 绑定
numpy: NumPy 数组操作
ndarray: Rust 多维数组
asr-data >= 0.1.1: 音频解码、波形处理与重采样
image >= 0.25.0: 图像处理
regex: 正则表达式

作者

wangmengdi - 790990241@qq.com

License

本仓库当前 未附带 LICENSE 文件；使用前请自行确认授权范围或与维护者联系。若在 PyPI 发布页有单独声明，以发布页为准。

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

osc_data-0.3.2.tar.gz (718.0 kB view details)

Uploaded Aug 11, 2026 Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

osc_data-0.3.2-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.9 MB view details)

Uploaded Aug 11, 2026 PyPymanylinux: glibc 2.17+ x86-64

osc_data-0.3.2-pp311-pypy311_pp73-manylinux_2_17_i686.manylinux2014_i686.whl (8.0 MB view details)

Uploaded Aug 11, 2026 PyPymanylinux: glibc 2.17+ i686

osc_data-0.3.2-cp314-cp314t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.9 MB view details)

Uploaded Aug 11, 2026 CPython 3.14tmanylinux: glibc 2.17+ x86-64

osc_data-0.3.2-cp314-cp314t-manylinux_2_17_i686.manylinux2014_i686.whl (8.0 MB view details)

Uploaded Aug 11, 2026 CPython 3.14tmanylinux: glibc 2.17+ i686

osc_data-0.3.2-cp314-cp314-win_amd64.whl (7.8 MB view details)

Uploaded Aug 11, 2026 CPython 3.14Windows x86-64

osc_data-0.3.2-cp314-cp314-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.9 MB view details)

Uploaded Aug 11, 2026 CPython 3.14manylinux: glibc 2.17+ x86-64

osc_data-0.3.2-cp314-cp314-manylinux_2_17_i686.manylinux2014_i686.whl (8.0 MB view details)

Uploaded Aug 11, 2026 CPython 3.14manylinux: glibc 2.17+ i686

osc_data-0.3.2-cp314-cp314-macosx_11_0_arm64.whl (7.0 MB view details)

Uploaded Aug 11, 2026 CPython 3.14macOS 11.0+ ARM64

osc_data-0.3.2-cp314-cp314-macosx_10_12_x86_64.whl (7.4 MB view details)

Uploaded Aug 11, 2026 CPython 3.14macOS 10.12+ x86-64

osc_data-0.3.2-cp313-cp313-win_amd64.whl (7.9 MB view details)

Uploaded Aug 11, 2026 CPython 3.13Windows x86-64

osc_data-0.3.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.9 MB view details)

Uploaded Aug 11, 2026 CPython 3.13manylinux: glibc 2.17+ x86-64

osc_data-0.3.2-cp313-cp313-manylinux_2_17_i686.manylinux2014_i686.whl (7.9 MB view details)

Uploaded Aug 11, 2026 CPython 3.13manylinux: glibc 2.17+ i686

osc_data-0.3.2-cp313-cp313-macosx_11_0_arm64.whl (7.0 MB view details)

Uploaded Aug 11, 2026 CPython 3.13macOS 11.0+ ARM64

osc_data-0.3.2-cp313-cp313-macosx_10_12_x86_64.whl (7.4 MB view details)

Uploaded Aug 11, 2026 CPython 3.13macOS 10.12+ x86-64

osc_data-0.3.2-cp312-cp312-win_amd64.whl (7.9 MB view details)

Uploaded Aug 11, 2026 CPython 3.12Windows x86-64

osc_data-0.3.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.9 MB view details)

Uploaded Aug 11, 2026 CPython 3.12manylinux: glibc 2.17+ x86-64

osc_data-0.3.2-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl (7.9 MB view details)

Uploaded Aug 11, 2026 CPython 3.12manylinux: glibc 2.17+ i686

osc_data-0.3.2-cp312-cp312-macosx_11_0_arm64.whl (7.0 MB view details)

Uploaded Aug 11, 2026 CPython 3.12macOS 11.0+ ARM64

osc_data-0.3.2-cp312-cp312-macosx_10_12_x86_64.whl (7.4 MB view details)

Uploaded Aug 11, 2026 CPython 3.12macOS 10.12+ x86-64

osc_data-0.3.2-cp311-cp311-win_amd64.whl (7.8 MB view details)

Uploaded Aug 11, 2026 CPython 3.11Windows x86-64

osc_data-0.3.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.9 MB view details)

Uploaded Aug 11, 2026 CPython 3.11manylinux: glibc 2.17+ x86-64

osc_data-0.3.2-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl (8.0 MB view details)

Uploaded Aug 11, 2026 CPython 3.11manylinux: glibc 2.17+ i686

osc_data-0.3.2-cp311-cp311-macosx_11_0_arm64.whl (7.0 MB view details)

Uploaded Aug 11, 2026 CPython 3.11macOS 11.0+ ARM64

osc_data-0.3.2-cp311-cp311-macosx_10_12_x86_64.whl (7.5 MB view details)

Uploaded Aug 11, 2026 CPython 3.11macOS 10.12+ x86-64

osc_data-0.3.2-cp310-cp310-win_amd64.whl (7.8 MB view details)

Uploaded Aug 11, 2026 CPython 3.10Windows x86-64

osc_data-0.3.2-cp310-cp310-win32.whl (6.7 MB view details)

Uploaded Aug 11, 2026 CPython 3.10Windows x86

osc_data-0.3.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.9 MB view details)

Uploaded Aug 11, 2026 CPython 3.10manylinux: glibc 2.17+ x86-64

osc_data-0.3.2-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl (8.0 MB view details)

Uploaded Aug 11, 2026 CPython 3.10manylinux: glibc 2.17+ i686

osc_data-0.3.2-cp310-cp310-macosx_11_0_arm64.whl (7.0 MB view details)

Uploaded Aug 11, 2026 CPython 3.10macOS 11.0+ ARM64

osc_data-0.3.2-cp310-cp310-macosx_10_12_x86_64.whl (7.5 MB view details)

Uploaded Aug 11, 2026 CPython 3.10macOS 10.12+ x86-64

osc_data-0.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.9 MB view details)

Uploaded Aug 11, 2026 CPython 3.9manylinux: glibc 2.17+ x86-64

osc_data-0.3.2-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl (8.0 MB view details)

Uploaded Aug 11, 2026 CPython 3.9manylinux: glibc 2.17+ i686

File details

Details for the file osc_data-0.3.2.tar.gz.

File metadata

Download URL: osc_data-0.3.2.tar.gz
Upload date: Aug 11, 2026
Size: 718.0 kB
Tags: Source
Uploaded using Trusted Publishing? No
Uploaded via: maturin/1.14.1

File hashes

Hashes for osc_data-0.3.2.tar.gz
Algorithm	Hash digest
SHA256	`021df892bb37c14503ef49d8075154b1b1299a3dc3cdc5c5cffd5ea0699e72b6`
MD5	`6938246d44e485f1d7dd24f84fc80e0a`
BLAKE2b-256	`f66e40ad1da8f5cb08a0817d8ece0f279dde25d835845e5bc1b5c81cf05efe5b`

Algorithm	Hash digest
SHA256	`6d5b90f7e380edbe509397283a43a2200f50c02e85fbfa2ec98d3f28d836f2a3`
MD5	`1e73f2a11003b9894d78774e1fa0e23e`
BLAKE2b-256	`bd6b15a04f6c1ef7865395627e91f56f37d0ded6113db550da904fe4f68393b6`

Algorithm	Hash digest
SHA256	`185eff8b0594c1442515c8b04064ce158b34d9976994b260d92bcbf072dc8bb0`
MD5	`fb80d85932502166335e23d3be264c10`
BLAKE2b-256	`ed6385a8028ebc8fcf504520f8159ad9ca7e4a297eb7392fa81b2489f4b93dab`

Algorithm	Hash digest
SHA256	`a756d9bb62abc01fe3170111f61c691ebf0a3e802d614be58be17a0d17a77dce`
MD5	`cfb3bccda0efa74e265c6a1fc31a6c47`
BLAKE2b-256	`c8b93733418b0aa3ae3aa569afaf99b886a92fdb0564620b286bf909ab218688`

Algorithm	Hash digest
SHA256	`9ef1ded5afd6d50de9ffeaaf86e7a76e5211e27802988be49261b88931de0b40`
MD5	`8cc3432fd2b1dea5fbe1779ba4f168f2`
BLAKE2b-256	`e401940317647c30fd2bfe702bb0a3207d39b738b5c28a2bd0b09a723e28f5fd`

Algorithm	Hash digest
SHA256	`3a7286c60b62edcef48a3152837b50f26c05cf0d88e899ce56e500462a8e4d99`
MD5	`a606431c9cba7b1fef768b71af3f8744`
BLAKE2b-256	`9bca3790a1f1cb37c6c6edd0c5f187f040b28e0575910d5fbb48248d2671a6a9`

Algorithm	Hash digest
SHA256	`75154f9619621c4f21efa5129a3c3f5a020efe9b283095c7b271b7bd33a35a65`
MD5	`5681dccbd2df0236bc176b2ade868ee4`
BLAKE2b-256	`aefcb8a70e9c9b9992708f355ba8da1f8aa70c2580bac7fa1d8b1fc1bcf4ea1b`

Algorithm	Hash digest
SHA256	`02cd0b8927c5c0bc4d6233b21c0d593ed483195ca7a6a52114f76b4e13fca5ef`
MD5	`728199118ecd5f185fbfe5297f2493fe`
BLAKE2b-256	`dabc4da909594a6615b661283ca07ac45412b2d3eb028ae0a5c12fd5881e3e4f`

Algorithm	Hash digest
SHA256	`b459a577a0ea8ea6ee433790796fc8b13b7375f64259cbfd401d942b48a84ced`
MD5	`7189eee7029abaeffe45316981f52956`
BLAKE2b-256	`ee1a603bacb00e8e95c41246ba5ac5e1968672f0e3402365730618cd19014e6e`

Algorithm	Hash digest
SHA256	`83c6b119746d369e42cd9122f914028eb032da6f52e7425419b3f0429f0620da`
MD5	`b444f1a752f8f2fb8e3d62c935845b02`
BLAKE2b-256	`b4dfbe19063a715b9e38b2e330174116b8f933ab92f220dc753b4f8d218c7de7`

Algorithm	Hash digest
SHA256	`61e853317e05a0e4add3a099e52d702cfbfb80963242c5fc26132fe26304deab`
MD5	`9e49db49cf48c85eff9b4d54e35afecf`
BLAKE2b-256	`8e44b06030ab877a8791ed6df67407a9ee39b6042dd729009df6d493f8352420`

Algorithm	Hash digest
SHA256	`3d74c36f0cb6a75222c1d19454568926e5afb389e6d1035de1b5783a963da449`
MD5	`719b0930ea9ec9b6669b8eb735e3cf06`
BLAKE2b-256	`2b69a53be5c26c875d780de2ded6579a223d4a68495824358a07714da97f4c98`

Algorithm	Hash digest
SHA256	`0f13e5ac4103b95110a71f73e4a5b49a880b301b8e89a98c89d7b6ddc85b3cf0`
MD5	`f7ed10ddfb1edb9c27b537384956f2e5`
BLAKE2b-256	`0e8ab4330bb82dbf0e03e6c2ccdae5a317a998d88d4f1b8965fc47be9d556b95`

Algorithm	Hash digest
SHA256	`25b10889ae0575aa4b0978a22f20e8f12b273d36c0e22e40018bd1aa299a5120`
MD5	`8d9d0cb8c185ad4870e8eee533a4f8c8`
BLAKE2b-256	`0d682f86aaf12364717bc9a7fd4e16cc4ed8e9834e8c36abad4c15964a7976fb`

Algorithm	Hash digest
SHA256	`8a85dd649d081f3c3d2a8c61a7cae037f5280c203b7fc9bdd9d1c08b010b5ed1`
MD5	`d747e2b57ad325d521b58abc803a7010`
BLAKE2b-256	`b7ffd6974e7523c44809b1a3f5bc1dd5a9eb896a8dda832ee4b73d37252919fc`

Algorithm	Hash digest
SHA256	`c207c3b07d60c2983b118907f4db8ee4d85e67bd2dd3fadf26c8a4908b19ba4f`
MD5	`f7b3cbd65a153d56077e4886e63649b2`
BLAKE2b-256	`7866b7fba4a0d44182da14168c6bd8ed9ba5ea0dde32167d1ad0c6740d21ded1`

Algorithm	Hash digest
SHA256	`31b9eb84fd1c938d939116ea0d31c4583d8a74afd459b9ab9c2684388d27b48c`
MD5	`117e351ccc0b56b2f1f901e1265ddb48`
BLAKE2b-256	`88b5f8a01ee94606543e8b2ad93c785d1dd10f042ec702b2cf67f5d9d15e9569`

Algorithm	Hash digest
SHA256	`f8e25d65d62a199699c44c39c205ff0745cafda86bb6090c9fe0df3f62b8e64e`
MD5	`12f9a884c88c776fcb91dadaa8cb49d0`
BLAKE2b-256	`28a792d3a60c05a62f58e655482103a72600fc4617b8e3eec23ba1e3013c4715`

Algorithm	Hash digest
SHA256	`724c7e998aa82fc070e2de91fdfa003268cc8160e8679022159631267e420794`
MD5	`e4473551886835f38670acac04246b2c`
BLAKE2b-256	`f84f717ae3c692424232509d592021ee30b8bcf702248ff9d01a5aeb9485e686`

Algorithm	Hash digest
SHA256	`7d1f1a04f30d2fe97b85940390f6a6957db546aba22b74ba0f3241a9bc9b4701`
MD5	`eb281b1932fd65a2edb36b95784b2c2d`
BLAKE2b-256	`871ae4e2a0636998f49ac1653d475470d1a81bc1d2c24cd878455fa332755ece`

Algorithm	Hash digest
SHA256	`06cc565ae68f51aa131b606903bbc237240ff085e2aeb209009f350d843dcd47`
MD5	`10dec46aae6bfd851fbb3e384a60b3cd`
BLAKE2b-256	`12f4307c6cabdc902afc0eb43291fada86e27d82129a62710270ea8675dcf7c8`

osc-data 0.3.2

osc-data

简介

特性

安装

环境要求

使用 pip 安装

从源码安装

快速开始

图像处理

文本处理

文本流处理

视频处理

音频处理

API 文档

Image 类

TextNormalizer 类

TextCleaner 类

TextStreamSentencizer 类

Audio 类

Video 类

项目结构

开发

构建 Rust 扩展

运行测试

代码风格

依赖

Python 依赖

Rust 依赖

作者

License

Key dates

1 maintainer

Credits

Requires

Download files

Source Distribution

Built Distributions

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

Algorithm	Hash digest
SHA256	`236b14220b22d6c2e1bff45dce4207c8f58bc32576bbac120bfdc116ef2f94c0`
MD5	`c859037cd3899d982fd15d607d4d7fef`
BLAKE2b-256	`668cd9a17ff14da3c53e74ebc2cc46e1893c8e6791c3522342005980fcd47acb`

Algorithm	Hash digest
SHA256	`7a045b5a6a480c9274606056ef40b5dfaeb15ca401b29b25f9921035be945caf`
MD5	`8db482fe2de1d974a59118fd9246986a`
BLAKE2b-256	`5b85ccaf9de0ff69304164c6a112c58f0e6231bec7e4d2819eeb5924e0023ec5`

Algorithm	Hash digest
SHA256	`70346804c87202d822be439524bfd7ebf2a781fce4fe5a742d58cf0a7901809f`
MD5	`e5e648d16f75de7b59bcb18af299827b`
BLAKE2b-256	`8861dd2a0b7d86ac626b7dc5bc1bf9c379a473c4117f16297e8df62d2a74bb29`

Algorithm	Hash digest
SHA256	`4a33cd268f99fd37fa829bddd02734640df6f7636889ed45d440ea6bd1bf4ee9`
MD5	`576eef51363131b31eefa1ba66997c65`
BLAKE2b-256	`3ec08b8270312d20b40f47a8f97c62dc3c8d9c3a9f6651e425d6126caf119646`

Algorithm	Hash digest
SHA256	`ae8af7ca911931dc173689dc31893d97fb806d4e0e37495cffb240fca52b612f`
MD5	`a61159da8a5367424be47a9cd222916d`
BLAKE2b-256	`20545fa8031a769896cf22541dc43f28e3d7a9faca11751b45ac67e5ab9d16c1`

Algorithm	Hash digest
SHA256	`8e1fba5313f1e20291fee4e2ee754e33467c805ff757c10f0cf61c70becef39f`
MD5	`d61bf1048802544767487484783d91ac`
BLAKE2b-256	`b1e52aeb075f9399136fe38ea8392b1a83e0b2b402af7d4939a96563effda99e`

Algorithm	Hash digest
SHA256	`cffc5b0792f2d03894adfb0267409443d9c41765dc85a9cede2a0feb1befb2aa`
MD5	`f8bc454b261341ffbd4d70624a90d58a`
BLAKE2b-256	`edad8fcc052f67ad2df7d1f9faa0cb9504bfdc4030f1d924e7d5a1137a0311cb`

Algorithm	Hash digest
SHA256	`adaa5139823dbc2cd9ffd41dce7fead00f3cdc89168a7e2c99b66924c80cfb42`
MD5	`7fa9a949250cb3e6788041ad2095b21f`
BLAKE2b-256	`c98e1a4b8daf8672f01770dd5fdd113a19bb8337ae43aa60e52abade29180ec3`

Algorithm	Hash digest
SHA256	`524699830bf2e1d7599abd929c8ac6bc07274c88ab195c333830f0299b242c27`
MD5	`7e4000f901b8a50eb7b49acb4e6cce3b`
BLAKE2b-256	`27159f4e4992b3ba769ee49a634d5e21b54084e0045e6873219fe0f2d5778f52`

Algorithm	Hash digest
SHA256	`5331db89e7c400a7d46aaa8a2633a8373897a9bfff694c3eab2c59a4507568aa`
MD5	`f3c476c34e3d47e19d41c91207d9203f`
BLAKE2b-256	`6dc9b59f33f8845902bea83a84af94d90911531a2c295e0faad6f2874639dd6e`