这是课程材料仓库,不是可直接安装的软件包,需按各 week 目录下 README 的说明操作;讲义录像主要为俄语,仅部分周有英文录像;作业与项目部分标注为 TBA,尚未提供
dla
markovka17MIT活跃维护一套面向音频处理的深度学习实战代码集,覆盖语音识别、说话人验证、关键词唤醒、语音合成与音色转换等任务,可直接参考实现来训练和微调模型。
Deep learning for audio processing
README 里没有找到可直接复制的安装命令,请到 README 或官方文档查看安装方式。
README 未说明
HSE 音频深度学习课程材料,按周提供讲义与研讨会代码,覆盖语音识别与音源分离
一套面向音频处理的深度学习实战代码集,覆盖语音识别、说话人验证、关键词唤醒、语音合成与音色转换等任务,可直接参考实现来训练和微调模型。
核心能力
- 按 week01 至 week06 分目录组织讲义与研讨会材料
- 覆盖 DSP、频谱图与 MFCC 等信号处理基础
- 讲解 CTC、DeepSpeech2、Conformer 与 Whisper 等 ASR 模型
- 介绍 wav2vec 2.0、HuBERT 等语音自监督学习方法
- 涉及 Demucs、DCCRN 等音源分离与降噪架构
容易踩的坑
暂无 release 数据(可能没有发版,或尚未同步)。
用户点评
登录用户的真实使用记录,可按版本和硬件筛选还没有人点评。第一个说出真实体验的人最有价值。

Deep Learning for Audio (DLA)
- Lecture and seminar materials for each week are in
./week*folders, seeREADME.mdfor materials and instructions - Any technical issues, ideas, bugs in course materials, contribution ideas - add an issue
- The current version of the course is conducted in autumn 2026 at the CS Faculty of HSE.
For previous years versions, see Past Versions section.
Syllabus
-
week01 Introduction to Course
- Lecture: Introduction to Course + Inspiration
- Seminar: Free talk
- Self-Study: Introduction to
PyTorchand basic devOps
-
week02 Introduction to Digital Signal Processing
- Lecture: Signals, Fourier Transform, spectrograms, MelScale, MFCC
- Seminar: DSP in practice, spectrogram creation, IRF, frequency filtering
-
week03 Automatic Speech Recognition I
- Lecture: Metrics, Datasets, Connectionist Temporal Classification (CTC), DeepSpeech2, Conformer, Beam Search, Language models
- Seminar: Audio Augmentations, WER and CER, CTC Decoding
-
week04 Automatic Speech Recognition II
- Lecture: LAS, Hybrid CTC/Attention, OpenAI Whisper, RNN-T, Streaming ASR, Decoder-only ASR
- Seminar: Whisper: greedy decoding, prompting, alignment in cross-attention, language forcing
-
week05 Automatic Speech Recognition III and Audio SSL
- Lecture: Self-Supervised Learning for Speech, wav2vec 2.0, HuBERT, BEST-RQ, GigaAM, LLM-based ASR, Delayed Streams Modeling, Encoder-free Architectures
-
week06 Source Separation I
- Lecture: A review of general Source Separation and Denoising, Encoder-Decoder-Separator architectures, Demucs family, DCCRN, FullSubNet+, BandSplitRNN
- Seminar: Metrics
TBA
Homeworks and Projects
TBA
See our project template.
Resources
Some of the weeks have English recordings. See the corresponding sub-directories.
Contributors & course staff
Course materials and teaching (in different years) were delivered by:
- Maxim Kaledin
- Georgy Gospodinov
- Georgiy Pistsov
- Aibek Alanov (previously)
- Petr Grinberg (previously)
- Grigory Fedorov(previously)
- Assel Yermekova (previously)
- Alexander Markovich (previously)
- Daniil Ivanov (previously)
- Ilya Lewin (previously)
- Timofey Smirnov (previously)
- Alexander Mamaev (previously)