Skip to content

  • 体验新版
    • 正在加载...
  • 登录
  • PaddlePaddle
  • models
  • Issue
  • #115

M
models
  • 项目概览

PaddlePaddle / models
接近 2 年 前同步成功

通知 230
Star 6828
Fork 2962
  • 代码
    • 文件
    • 提交
    • 分支
    • Tags
    • 贡献者
    • 分支图
    • Diff
  • Issue 602
    • 列表
    • 看板
    • 标记
    • 里程碑
  • 合并请求 255
  • Wiki 0
    • Wiki
  • 分析
    • 仓库
    • DevOps
  • 项目成员
  • Pages
M
models
  • 项目概览
    • 项目概览
    • 详情
    • 发布
  • 仓库
    • 仓库
    • 文件
    • 提交
    • 分支
    • 标签
    • 贡献者
    • 分支图
    • 比较
  • Issue 602
    • Issue 602
    • 列表
    • 看板
    • 标记
    • 里程碑
  • 合并请求 255
    • 合并请求 255
  • Pages
  • 分析
    • 分析
    • 仓库分析
    • DevOps
  • Wiki 0
    • Wiki
  • 成员
    • 成员
  • 收起侧边栏
  • 动态
  • 分支图
  • 创建新Issue
  • 提交
  • Issue看板
已关闭
开放中
Opened 6月 21, 2017 by saxon_zh@saxon_zhGuest

Pruning speedups beam search

Created by: kuke

The ctc beam search in DS2, or prefix beam search consists of appending candidate characters to prefixes and repeatedly looking up the n-gram language model. Both the two processes are time-consuming, and bring difficulty in parameters tuning and deployment.

A proven effective way is to prune the beam search. Specifically, in extending prefix only the fewest number of characters whose cumulative probability is at least p need to be considered, instead of all the characters in the vocabulary. When p is taken 0.99 as recommended by the DS2 paper, about 20x speedup is yielded in English transcription than that without pruning, with very little loss in accuracy. And for the Mandarin, the speedup ratio is reported to be up to 150x.

Due to pruning, the tuning of parameters gets more efficiently. There are two important parameters alpha and beta in beam search, associated with language model and word insertion respectively. With a more acceptable speed, alpha and beta can be searched elaborately. And the relation between WER and the two parameters turns out to be: figure_1

With the optimal parameters alpha=0.26 and beta=0.1 as shown in above figure, currently the beam search decoder has decreased WER to 13% from the best path decoding accuracy 22%.

指派人
分配到
无
里程碑
无
分配里程碑
工时统计
无
截止日期
无
标识: paddlepaddle/models#115
渝ICP备2023009037号

京公网安备11010502055752号

网络110报警服务 Powered by GitLab CE v13.7
开源知识
Git 入门 Pro Git 电子书 在线学 Git
Markdown 基础入门 IT 技术知识开源图谱
帮助
使用手册 反馈建议 博客
《GitCode 隐私声明》 《GitCode 服务条款》 关于GitCode
Powered by GitLab CE v13.7