Search results

The generated audio length is one second shorter

Status: Open.

#50 In descriptinc/melgan-neurips;

· hahahahaha20250325 opened on Dec 22, 2024

For Ready-to-use env.yaml for training. (Create new file <<melgan3.6_env.yaml>> and paste this text, Required python 3.6)

Status: Open.

#49 In descriptinc/melgan-neurips;

· iamshreeji-copy1 opened on Dec 19, 2024

[CONTRIBUTION] Speech Dataset Generator

Status: Open.

#48 In descriptinc/melgan-neurips;

· davidmartinrius opened on Feb 29, 2024

How to normalize mel spectorgram extracted by Audio2Wav class?

Status: Open.

#45 In descriptinc/melgan-neurips;

· predawnang opened on May 10, 2023

How to measure the quality of synthetic audio with PESQ

Status: Open.

#44 In descriptinc/melgan-neurips;

· predawnang opened on May 5, 2023

Why is the generator loss continuously increasing instead of decreasing? Is continuous increase correct?

Status: Open.

#43 In descriptinc/melgan-neurips;

· jdwang125 opened on Mar 27, 2023

Why is D_fake calculated in "Train Generator" while D_real is calculated in "Train Discriminator"?

Status: Open.

#42 In descriptinc/melgan-neurips;

· jdwang125 opened on Mar 27, 2023

Melgan multispaeker pretrained model fine-tuning with single speaker

Status: Open.

#39 In descriptinc/melgan-neurips;

· MuruganR96 opened on Sep 29, 2022

VQ-VAE Implementation

Status: Open.

#38 In descriptinc/melgan-neurips;

· dylanprins opened on Feb 7, 2022

what is the learning rate should be when finetune a small dataset on a base model

Status: Open.

#37 In descriptinc/melgan-neurips;

· Liujingxiu23 opened on Jan 26, 2021

Why perform Audio2Mel's method on extracting mel spectrogram?

Status: Open.

#36 In descriptinc/melgan-neurips;

· shawnbzhang opened on Nov 24, 2020

Problem in changing sample rate into 16k Hz

Status: Open.

#35 In descriptinc/melgan-neurips;

· xuexidi opened on Nov 16, 2020