abhishekyana Goto Github PK

speech-to-text-wavenet

Speech-to-Text-WaveNet : End-to-end sentence level English speech recognition based on DeepMind's WaveNet and tensorflow

sqlitedict

Persistent dict, backed by sqlite3 and pickle, multithread-safe.

supervised-image-classification

Image classification, Multi Label and Multi Class object detection using CaffeNet and VGG models trained from scratch on PASCAL 2007 dataset, onto 20classes. Analysed the results with tSNE and KNN plots.

tecogan

This repo will contain source code and materials for the TecoGAN project, i.e. code for a TEmporally COherent GAN

tensorflow

Open source software library for numerical computation using data flow graphs.

tensorflow-examples

TensorFlow Tutorial and Examples for Beginners with Latest APIs

text-to-text-transfer-transformer

Code for the paper "Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer"

torch7

http://torch.ch

unpaired-image-to-image-translation-using-cycle-consistent-adversarial-networks

Implemented a Paper applied to convert the Natural Image to Anime style Image

unsupervised-image-localization-with-heatmaps

Weakly Supervised Image Localization

utilsroutine

UtilsRoutine

visual-question-answering-with-coattention-mechanism

Visual Question Answering System, Trained on MS COCO dataset, Implemented SimpleBaseLineNet and A Variant of Co-Attention Mechanism from a Paper

voice-fingerprinting

Voice Fingerprinting using deep learning, wide range of applications from personalised searches to unlocking machines etc

voice_gender_recognition

webapps

WebApps I developed to make my work easier.

whatsappchatparser

This helps in parsing the txt file exported from WhatsApp. This txt file is converted into csv file which is will help in analysing data easily with ML techniques

word-embeddings-with-dash-architecture

Word Embeddings with DASH architecture

zeronn_asimo

This is an AI(artificial Intelligence) Assistant that uses Neural Networks for the speech to text conversion and trained using my voice dataset and Neural Networks used are called LSTM recurrent Neural Networks. The converted Text is then analysed using the Natural Language Processing NLTK and tokenizes the sentences and words and triggers the appropriate actions based on the user rules. Then based on the actions done the AI then makes a phrase and that is converted into a speech using the Text-to-Speech algorithm. So, TO sum it up the system takes the users queries as voice inputs and processes that and does the tasks and gives the output as sound and actions. This processing is a powerful algorithm that uses the free available data online than the costly APIs.

zeta_beta

This APP is purely based on java from scratch. This app is used in parallel with the Fingerprint Recognition Project made using Fingerprint sensor module and BLE(Bluetooth Low Energy) module HC-10. Connects with the Android phone.

zframework

Z Framework for Easy Deep Learning