Automatic Speech Recognition (ASR)

Automatic Speech Recognition (ASR) involves converting spoken language into written text. It is designed to transcribe spoken words into text in real-time, allowing people to communicate with computers, mobile devices, and other technology using their voice. The goal of Automatic Speech Recognition is to accurately transcribe speech, taking into account variations in accent, pronunciation, and speaking style, as well as background noise and other factors that can affect speech quality.

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 621–630 of 3012 papers

Title	Date	Tasks	Status
Romanization Encoding For Multilingual ASR	Jul 5, 2024	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	—Unverified
Seed-ASR: Understanding Diverse Speech and Contexts with LLM-based Speech Recognition	Jul 5, 2024	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	—Unverified
Written Term Detection Improves Spoken Term Detection	Jul 5, 2024	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	CodeCode Available
Semi-supervised Learning for Code-Switching ASR with Large Language Model Filter	Jul 5, 2024	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	—Unverified
Improving Accented Speech Recognition using Data Augmentation based on Unsupervised Text-to-Speech Synthesis	Jul 4, 2024	Accented Speech RecognitionAutomatic Speech Recognition	—Unverified
Multi-Convformer: Extending Conformer with Multiple Convolution Kernels	Jul 4, 2024	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	—Unverified
Codec-ASR: Training Performant Automatic Speech Recognition Systems with Discrete Speech Representations	Jul 3, 2024	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	—Unverified
Error Correction by Paying Attention to Both Acoustic and Confidence References for Automatic Speech Recognition	Jun 29, 2024	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	—Unverified
Enhanced ASR Robustness to Packet Loss with a Front-End Adaptation Network	Jun 27, 2024	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	CodeCode Available
Tradition or Innovation: A Comparison of Modern ASR Methods for Forced Alignment	Jun 27, 2024	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	—Unverified

Show:10 25 50

← PrevPage 63 of 302Next →

All datasets LRS2 RealMAN Sagalee HUI speech corpus LRS3-TED M-AILabs speech dataset The Spoken Wikipedia Corpora Voxforge German VoxPopuli

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	TM-CTC	Test WER	10.1	—	Unverified
2	TM-seq2seq	Test WER	9.7	—	Unverified
3	CTC/attention	Test WER	8.2	—	Unverified
4	LF-MMI TDNN	Test WER	6.7	—	Unverified
5	Whisper-LLaMA	Test WER	6.6	—	Unverified
6	End2end Conformer	Test WER	3.9	—	Unverified
7	End2end Conformer	Test WER	3.7	—	Unverified
8	MoCo + wav2vec (w/o extLM)	Test WER	2.7	—	Unverified
9	CTC/Attention	Test WER	1.5	—	Unverified
10	Whisper	Test WER	1.3	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SpatialNet	CER	14.5	—	Unverified
2	CleanMel-L-mask	CER	14.4	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Conformer	Test WER	15.32	—	Unverified
2	Whisper-largev3-finetuned	Test WER	10.82	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Conformer Transducer	WER (%)	1.89	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DistillAV	WER	1.4	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Conformer Transducer	WER (%)	4.28	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Conformer Transducer	WER (%)	8.04	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Conformer Transducer	WER (%)	3.36	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Conformer Transducer (German)	WER (%)	8.98	—	Unverified