Word Error Rate (WER): 56.6
Papers with Code · CC BY-SA 4.0
Speech Recognition
Speech Recognition is the task of converting spoken language into text. It involves recognizing the words spoken in an audio recording and transcribing them into a written format. The goal is to accurately transcribe the speech in real-time or from recorded audio, taking into account factors such as accents, speaking speed, and background noise.
<span style="color:grey; opacity: 0.6">( Image credit: SpecAugment )</span>
개요는 각 항목의 일부입니다. 위 탭에서 전체 기록을 페이지별로 탐색할 수 있습니다.
벤치마크
Papers with Code 아카이브의 데이터셋별 기록 예시입니다. 평가 조건과 지표 방향이 검증되지 않아 순위를 의미하지 않습니다.
Dataset기록된 모델 예시MetricPaperCode
AudioShake v3
Word Error Rate (WER): 17.3Punctuation F-1: 65.3Parenthesis F-1: 37.9Line break F-1: 84.3
논문
코드
AudioShake v3
Word Error Rate (WER): 16.1Punctuation F1: 57.0Line break F1: 84.4Section break F1: 73.9
논문
코드
AudioShake v3
Word Error Rate (WER): 20.8Punctuation F-1: 46.1Parenthesis F-1: 3.2Line break F-1: 88.6
논문
코드
AudioShake v3
Word Error Rate (WER): 12.6Punctuation F-1: 57.1Parenthesis F-1: 76.6Line break F-1: 83.7
논문
코드
데이터셋
292개논문
Word Error Rate (WER): 5.7
Word Error Rate (WER): 6.23
Word Error Rate (WER): 5.68
ABX-within: 12.05ABX-across: 20.6
Word Error Rate (WER): 1.4
Word Error Rate (WER): 0.81
Word Error Rate (WER): 1.3
Word Error Rate (WER): 5.0
Word Error Rate (WER): 0.68
라이브러리
Papers with Code 리더보드 row에 연결된 구현 저장소입니다.
하위 태스크
AL
Automatic Lyrics Transcription
AS
Accented Speech Recognition
AS
Automatic Speech Recognition (ASR)
VS
Visual Speech Recognition
NS
Noisy Speech Recognition
DS
Distant Speech Recognition
EC
English Conversational Speech Recognition
RS
Robust Speech Recognition
SS
Sequence-To-Sequence Speech Recognition
SL
Speech Language Identification
TS
Target Speaker Extraction