2 个仓库
Binary representations of datasets optimized for high-speed loading during model training.
Distinct from Dataset Formats: Specifically covers the conversion of text to binary for training efficiency, not temporal sequence formatting.
Explore 2 awesome GitHub repositories matching data & databases · Binarized. Refine with filters or upvote what's useful.
Fairseq is a PyTorch toolkit for sequence-to-sequence modeling, specializing in neural machine translation, automatic speech recognition, and large-scale language model training. It provides a framework for processing and aligning diverse data sources, including text, audio, and video, to support tasks such as speech-to-text conversion and multimodal sequence learning. The project is distinguished by its distributed training capabilities, which utilize parameter sharding, mixed-precision training, and CPU offloading to handle models that exceed single-device memory. It also includes specializ
Processes raw text and alignment files into a binary format for efficient loading during training.
Muzic 是一个用于 AI 驱动的音乐分析、创作和合成的深度学习平台和框架。它作为一个音乐生成框架和分析工具,利用大型语言模型和自主智能体来编排符号音乐和音频音乐的创作与解读。 该项目以其跨模态能力而著称,将自然语言和符号音乐映射到共享的联合嵌入空间中,用于零样本分类和信息检索。它采用了多种专门的架构,包括用于音频合成的扩散框架、用于长序列结构一致性的双粒度注意力机制,以及结合音乐理论规则与神经网络的混合系统。 该平台涵盖了广泛的功能,包括从文本和歌词生成 MIDI 序列、神经歌声合成以及自动歌词转录。它还提供用于音乐结构建模、基于属性的符号生成以及通过自主智能体编排外部音乐工具的工具。 支持性实用程序包括用于大规模 MIDI 二进制化、数据集编码的数据工程流水线,以及用于旋律音符提取和语音到音素对齐的音频信号处理。
Transforms raw MIDI data into specialized binarized formats to optimize large-scale model training and inference.