Vladislav Klim

RNNoise_Wrapper. A simple Python wrapper for audio noise reduction RNNoise. Simplifies work with it, adds new trained models and detailed instructions for training.

Voice_ChatBot. Chatbot in russian with speech recognition using PocketSphinx and speech synthesis using RHVoice. The AttentionSeq2Seq model is used. Imlemented using Python3+TensorFlow+Keras.

StressRNN. Modified version of RusStress (https://github.com/MashaPo/russtress) — python package for placing stress in Russian text using RNN (BiLSTM) and the "Grammatical Dictionary" by A. A. Zaliznyak (from http://odict.ru/).

Russian_subtitles_dataset. Preprocessing of the dataset of 347 subtitles for the TV series (thanks to Taiga Corpus) to build a word2vec model, JamSpell model, neural network training, chat bot training or in any other NLP task.

docker_image_with_cuda10_cudnn7. Dockerfiles and manual for easy build of docker image with CUDA10.X and cuDNN7.6 to run TensorFlow/PyTorch on the nvidia GPU in docker-container.

Uk_Stemmer. A small modification of the stemmer for the Ukrainian language (https://github.com/Amice13/ukr_stemmer)

WebRTCVAD_Wrapper. A simple Python wrapper to simplify working with WebRTC VAD and its rougher analogue based on RMS and ZCR (useful for processing audio recordings before using them with neural networks).

Simple_speech_synthesizer. A simple speech synthesizer based on phonemes. 4th laboratory work on the discipline EAzIIS (natural language interfaces intelligent systems) of the 7th semester of Department IIT BSUIR.

NN_Server. Back-end of the project of intelligence system of neural networks at the department of IIT BSUIR. Implemented with using Python3+TensorFlow+Keras.

building_tensorflow. Building TensorFlow v1.15 from source (for CPU and GPU, Python 3.7).