Mohamed Mesto's Projects
Experiment framework for performing DAT and MTL for accented speech recognition
a deep accent recognition network
PyTorch android examples of usage in applications
Architect crossword EN at Technical University of Berlin - ZEMS: Zentraleinrichtung Moderne Sprachen
Analysis and investigating the confounding effect of accents in end-to-end Automatic Speech Recognition models.
Accented speech Recognition (ASR) Accent Analysis for German Accents
:books: List of awesome university courses for learning Computer Science!
Video Streaming Mixer Library: In order to create e.g. playlists consisting of multiple existing different video streams it is needed to make sure that the different video stream sources are all compatible in regards to video resolution and bitrate and can be combined into a single video stream.
Car agency Management System using Oracle UI and DB
Care Robot Video in German - Pflegeroboter Video
Care Robot an episode on my youtube channel #Tips From Europe
Abstract—Increased number of technologies which use the same unlicensed frequency spectrum often cause severe degradation in the performance of these technologies. In this paper we will discuss the technologies that use the 5GHz frequency band. More and more participating technologies in the 5GHz band cause interference with each other which causes the corresponding technologies to face lower performance in their normal operation. We will discuss the possible challenges we face and their solutions in the coordination of WiFi operating at different frequencies e.g WiFi at 2.4GHz and WiFi at 5GHz. We further discuss the challenges and their solutions in the coexistence of WiFi at 5GHz with LTE-U and LAA. Wherein we will discuss also the deployment of LTE technologies coexisting with WiFi technologies in the unlicensed spectrum 5 GHz. Listen before talk (LBT) functionality is required in unlicensed band in order to ensure the fair coexistence among different operators. We propose a LBT enhancement algorithm with contention window size adaptation for LTE with Licensed-Assisted Access (LTE-LAA) in order to achieve not only channel access fairness but also the QoS fairness. In the last part of the paper we explore further to discuss the cross technology communication which enables distinct participating technologies to effectively communicate with each other to optimize the performance.
Common Voice is part of Mozilla's initiative to help teach machines how real people speak.
Stores documents used by the TensorFlow developer community
CrossWord Schneiderlein in German
State-of-the-Art Deep Learning scripts organized by models - easy to train and deploy with reproducible accuracy and performance on enterprise-grade infrastructure.
The code enables users to use Mozilla's Deep Speech model over the Web Browser.
Speech Recognition using DeepSpeech2.
TensorFlow documentation
🤗 Evaluate: A library for easily evaluating machine learning models and datasets.
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.
Transformer related optimization, including BERT, GPT
Fog Computing - Fog Store - IPFS vs GDP