Speech Recognition
Kfdhfljl
这个作者很懒,什么都没留下…
展开
-
ESPNET 环境配置
ESPNET 环境配置目录ESPNET 环境配置1、安装Ubuntu软件2、setting of the environment for GPU support:4.安装espnet1、安装Ubuntu软件基本环境:Python 3.6.1+(推荐3.7.3)gcc 4.9+ for PyTorch1.0.0+Optionally, GPU environment requires the following libraries:Cuda 8.0, 9.0, 9.1, 10.0 depe原创 2020-07-27 18:06:57 · 1553 阅读 · 1 评论 -
Efficient Active Learning for Automatic Speech Recognition... 论文笔记
Efficient Active Learning for Automatic Speech Recognition via Augmented Consistency Regularization,submited to INTERSPEECH2020文章链接目录Efficient Active Learning for Automatic Speech Recognition via Augmented Consistency Regularization,submited to INTERSPEE原创 2020-07-21 15:05:21 · 293 阅读 · 0 评论 -
active learning for speech recognition
active learning for speech recognition概述一些语音识别领域的主动学习方法Semi-supervised and active-learning scenarios: Efficient acoustic model refinement for a low resource Indian language, INTERSPEECH 2018使用的任务模型是:DNN-HMM framework trained using Kaldi and with word le原创 2020-07-20 18:03:24 · 212 阅读 · 0 评论 -
Active Learning Methods for Low Resource End-to-End Speech Recognition 论文笔记
Active Learning Methods for Low Resource End-To-End Speech Recognition,INTERSPEECH2019文章链接背景及概述a active learning method using a joint score of uncertainty and i-vector diversity.本文将主动学习应用于端到端(E2E)的语音识别。E2E模型使用encoder-decoder神经网络直接输出字词。相比于传统模型,E2E模型的计算原创 2020-07-20 11:31:44 · 288 阅读 · 0 评论