Unofficial implement with paper SpeakerGAN: Speaker identification with conditional generative adversarial network

Overview

Introduction

This repository is about paper SpeakerGAN , and is unofficially implemented by Mingming Huang ([email protected]), Tiezheng Wang ([email protected]) and thanks for advice from TongFeng.

SpeakerGAN paper

SpeakerGAN: Speaker identification with conditional generative adversarial network, by Liyang Chen , Yifeng Liu , Wendong Xiao , Yingxue Wang ,Haiyong Xie.

Usage

For train / test / generate:

python speakergan.py

You may need to change the path of wav vad preprocessed files.

Our results

acc: 94.27% with random sampled testset. 

acc: 93.21% with fixed start sampled testset.

using model file: model/49_D.pkl

acc: 98.44% on training classification accuracy with real samples.

There is about 4% gap on testset lower compared to paper result. We can't find out the reason. We want your help !

Alt accuracy
Alt loss_d_loss_g
Alt learning_rate

Details of paper

The following are details about this paper.

================ input ==================

  1. feature: fbank, 8000hz, 25ms frame, 10ms overlap. shape:(160,64)

  2. dataset: librispeech-100 train-clean-100 POI:251

  3. data preprocess: vad、mean and variance normalization, shuffled.

  4. 60% train. 40% test.

================ model architecture ==================

  1. dataflow: data -> feature extraction -> G & D

  2. model architecture:

    G: gated CNN, encoder-decoder, Huber loss + adversarial loss

    D: ResnetBlocks, template average pooling, FC, softmax, crossentropy loss + adversarial loss

  3. G: shuffler layer, GLU

  4. D: ReLU

================ training ==================

  1. lr: 0-9, 0.0005 | 9-49, 0.0002

  2. L(d): λ1 λ2 = 1

  3. batch_size: 64

  4. D_train steps / G_train steps = 4

  5. Ladv Loss: Label smoothing, 1 -> 0.7 ~ 1.0, 0 -> 0 ~ 0.3

======== not sure or differences with paper ========

  1. weights,bias initialize function, use: xavier_uniform and zeros

  2. pytorch huber_loss: + 0.5 to be same with paper. but no implement here.

  3. for shorter wav, paper: padded with zero. we: padded with feature again.

  4. gated cnn architecture.

  5. we use webrtcvad mode(3) for vad preprocess.

PyTorch implementation of the WarpedGANSpace: Finding non-linear RBF paths in GAN latent space (ICCV 2021)

Authors official PyTorch implementation of the "WarpedGANSpace: Finding non-linear RBF paths in GAN latent space" [ICCV 2021].

Christos Tzelepis 100 Dec 06, 2022
4th place solution for the SIGIR 2021 challenge.

SIGIR-2021 (Tinkoff.AI) How to start Download train and test data: https://sigir-ecom.github.io/data-task.html Place it under sigir-2021/data/. Run py

Tinkoff.AI 4 Jul 01, 2022
This repository contains an overview of important follow-up works based on the original Vision Transformer (ViT) by Google.

This repository contains an overview of important follow-up works based on the original Vision Transformer (ViT) by Google.

75 Dec 02, 2022
This repo contains code to reproduce all experiments in Equivariant Neural Rendering

Equivariant Neural Rendering This repo contains code to reproduce all experiments in Equivariant Neural Rendering by E. Dupont, M. A. Bautista, A. Col

Apple 83 Nov 16, 2022
CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection

CLOCs is a novel Camera-LiDAR Object Candidates fusion network. It provides a low-complexity multi-modal fusion framework that improves the performance of single-modality detectors. CLOCs operates on

Su Pang 254 Dec 16, 2022
Pytorch Lightning code guideline for conferences

Deep learning project seed Use this seed to start new deep learning / ML projects. Built in setup.py Built in requirements Examples with MNIST Badges

Pytorch Lightning 1k Jan 02, 2023
DeepMetaHandles: Learning Deformation Meta-Handles of 3D Meshes with Biharmonic Coordinates

DeepMetaHandles (CVPR2021 Oral) [paper] [animations] DeepMetaHandles is a shape deformation technique. It learns a set of meta-handles for each given

Liu Minghua 73 Dec 15, 2022
[ECCV'20] Convolutional Occupancy Networks

Convolutional Occupancy Networks Paper | Supplementary | Video | Teaser Video | Project Page | Blog Post This repository contains the implementation o

622 Dec 30, 2022
Article Reranking by Memory-enhanced Key Sentence Matching for Detecting Previously Fact-checked Claims.

MTM This is the official repository of the paper: Article Reranking by Memory-enhanced Key Sentence Matching for Detecting Previously Fact-checked Cla

ICTMCG 13 Sep 17, 2022
The official GitHub repository for the Argoverse 2 dataset.

Argoverse 2 API Official GitHub repository for the Argoverse 2 family of datasets. If you have any questions or run into any problems with either the

Argo AI 156 Dec 23, 2022
Pixel-wise segmentation on VOC2012 dataset using pytorch.

PiWiSe Pixel-wise segmentation on the VOC2012 dataset using pytorch. FCN SegNet PSPNet UNet RefineNet For a more complete implementation of segmentati

Bodo Kaiser 378 Dec 30, 2022
Robust Instance Segmentation through Reasoning about Multi-Object Occlusion [CVPR 2021]

Robust Instance Segmentation through Reasoning about Multi-Object Occlusion [CVPR 2021] Abstract Analyzing complex scenes with DNN is a challenging ta

Irene Yuan 24 Jun 27, 2022
I3-master-layout - Simple master and stack layout script

Simple master and stack layout script | ------ | ----- | | | | | Ma

Tobias S 18 Dec 05, 2022
Kaggle | 9th place (part of) solution for the Bristol-Myers Squibb – Molecular Translation challenge

Part of the 9th place solution for the Bristol-Myers Squibb – Molecular Translation challenge translating images containing chemical structures into I

Erdene-Ochir Tuguldur 22 Nov 30, 2022
Source code for our paper "Molecular Mechanics-Driven Graph Neural Network with Multiplex Graph for Molecular Structures"

Molecular Mechanics-Driven Graph Neural Network with Multiplex Graph for Molecular Structures Code for the Multiplex Molecular Graph Neural Network (M

shzhang 59 Dec 10, 2022
PyTorch code for the ICCV'21 paper: "Always Be Dreaming: A New Approach for Class-Incremental Learning"

Always Be Dreaming: A New Approach for Data-Free Class-Incremental Learning PyTorch code for the ICCV 2021 paper: Always Be Dreaming: A New Approach f

49 Dec 21, 2022
An implementation of the AlphaZero algorithm for Gomoku (also called Gobang or Five in a Row)

AlphaZero-Gomoku This is an implementation of the AlphaZero algorithm for playing the simple board game Gomoku (also called Gobang or Five in a Row) f

Junxiao Song 2.8k Dec 26, 2022
Official implementation of "A Unified Objective for Novel Class Discovery", ICCV2021 (Oral)

A Unified Objective for Novel Class Discovery This is the official repository for the paper: A Unified Objective for Novel Class Discovery Enrico Fini

Enrico Fini 118 Dec 26, 2022
RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching

RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching This repository contains the source code for our paper: RAFT-Stereo: Multilevel

Princeton Vision & Learning Lab 328 Jan 09, 2023
make ASCII Art by Deep Learning

DeepAA This is convolutional neural networks generating ASCII art. This repository is under construction. This work is accepted by NIPS 2017 Workshop,

OsciiArt 1.4k Dec 28, 2022