ASR evaluation

本文深入探讨了自动语音识别(ASR)系统的评价方法,重点介绍了转录准确性和错误率指标,如WER(Word Error Rate)和SER(Sentence Error Rate),并通过使用NIST Scoring Toolkit-SCLITE工具进行详细分析。

摘要生成于 C知道 ,由 DeepSeek-R1 满血版支持, 前往体验 >

Transcription accuracy
  • WER(Word Error Rate)
  • SER(Sentence Error Rate)
Semantic accuracy
Concept Accuracy(based on domain concepts)

Reference:

Tool:

SCTK has at its core, sclite (Score-Lite), which is a flexible Dynamic Programming alignment engine used to “align” errorful hypothesized texts, such as output from an ASR system, to the correct reference texts. After alignment, sclite generates a veriety of summary as well as detailed scoring reports.

评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值