TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

SotA/Audio/Speech Recognition

Speech Recognition

182 benchmarks6433 papers

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>

Benchmarks

Speech Recognition on LibriSpeech test-clean

Word Error Rate (WER)

Speech Recognition on LibriSpeech test-other

Word Error Rate (WER)

Speech Recognition on Switchboard + Hub500

Percentage error

Speech Recognition on TIMIT

Percentage error

Speech Recognition on AISHELL-1

Word Error Rate (WER)Params(M)

Speech Recognition on Jam-ALT English

Word Error Rate (WER)Punctuation F-1Line break F-1Case-Sensitive Word Error RateSection break F-1Case Error RateParenthesis F-1

Speech Recognition on WSJ eval92

Word Error Rate (WER)

Speech Recognition on Jam-ALT

Word Error Rate (WER)Punctuation F1Line break F1Case-Sensitive Word Error RateSection break F1Case Error RateParenthesis F-1

Speech Recognition on Jam-ALT French

Word Error Rate (WER)Punctuation F-1Line break F-1Case-Sensitive Word Error RateCase Error RateParenthesis F-1Section break F-1

Speech Recognition on Jam-ALT German

Word Error Rate (WER)Punctuation F-1Line break F-1Case-Sensitive Word Error RateCase Error RateSection break F-1Parenthesis F-1

Speech Recognition on Jam-ALT Spanish

Word Error Rate (WER)Punctuation F-1Line break F-1Case-Sensitive Word Error RateSection break F-1Case Error RateParenthesis F-1

Speech Recognition on LibriTTS

PESQPeriodicityV/UV F1M-STFTMCD

Speech Recognition on IndicTTS

Mean Opinion Score

Speech Recognition on swb_hub_500 WER fullSWBCH

Percentage error

Speech Recognition on LRS2

Test WERWord Error Rate (WER)

Speech Recognition on LRS3-TED

Word Error Rate (WER)WER

Speech Recognition on MediaSpeech

WER for SpanishWER for FrenchWER for ArabicWER for Turkish

Speech Recognition on SLUE

VoxPopuli (Dev)VoxPopuli (Test)VoxCeleb (Dev)VoxCeleb (Test)

Speech Recognition on VietMed

Dev WERTest WER

Speech Recognition on WenetSpeech

Character Error Rate (CER)

Speech Recognition on North American English

Mean Opinion Score

Speech Recognition on CHiME real

Percentage error

Speech Recognition on EasyCom

WER (%)

Speech Recognition on GigaSpeech DEV

Word Error Rate (WER)

Speech Recognition on GigaSpeech TEST

Word Error Rate (WER)

Speech Recognition on CHiME-6 dev_gss12

Word Error Rate (WER)

Speech Recognition on Hub5'00 SwitchBoard

SwitchBoardCallHomeEval2000Hub5'00

Speech Recognition on LJSpeech

Mean Opinion Score

Speech Recognition on Tedlium

Word Error Rate (WER)

Speech Recognition on WSJ dev93

Word Error Rate (WER)

Speech Recognition on CHiME-6 eval

Word Error Rate (WER)

Speech Recognition on Common Voice vi

Test WER

Speech Recognition on DIRHA English WSJ

Word Error Rate (WER)

Speech Recognition on Europarl-ASR EN Guest-test

WER

Speech Recognition on Fongbe audio

Word Error Rate (WER)

Speech Recognition on Libri-Light test-clean

Word Error Rate (WER)ABX-acrossABX-within

Speech Recognition on Libri-Light test-other

Word Error Rate (WER)ABX-acrossABX-within

Speech Recognition on Mandarin Chinese

Mean Opinion Score

Speech Recognition on SPGISpeech

Word Error Rate (WER)

Speech Recognition on Speech Commands

Accuracy (%)

Speech Recognition on VIVOS

Test WER

Speech Recognition on WSJ eval93

Word Error Rate (WER)

Speech Recognition on AISHELL-2

Word Error Rate (WER)

Speech Recognition on AMI IMH

Word Error Rate (WER)

Speech Recognition on AMI SDM1

Word Error Rate (WER)

Speech Recognition on Blizzard Challenge 2013

NLL

Speech Recognition on CHiME clean

Percentage error

Speech Recognition on CHiME-4 real 6ch

Word Error Rate (WER)

Speech Recognition on Europarl-ASR EN MEP-test

WER

Speech Recognition on GRID corpus (mixed-speech)

ESTOIPESQSTOI

Speech Recognition on LibriCSS

Word Error Rate (WER)

Speech Recognition on Lip2Wav (Chem)

ESTOIPESQSTOI

Speech Recognition on Lip2Wav (Chess)

ESTOIPESQSTOI

Speech Recognition on Lip2Wav (DL)

ESTOIPESQSTOI

Speech Recognition on Lip2Wav (EH)

ESTOIPESQSTOI

Speech Recognition on Lip2Wav (HS)

ESTOIPESQSTOI

Speech Recognition on RealMAN

CER

Speech Recognition on Sagalee

Test WER

Speech Recognition on TED-LIUM

Word Error Rate (WER)

Speech Recognition on VoxForge American-Canadian

Percentage error

Speech Recognition on VoxForge Commonwealth

Percentage error

Speech Recognition on VoxForge European

Percentage error

Speech Recognition on VoxForge Indian

Percentage error

Speech Recognition on AISHELL-2 Test Android

Word Error Rate (WER)

Speech Recognition on AISHELL-2 Test IOS

Word Error Rate (WER)

Speech Recognition on AISHELL-2 Test Mic

Word Error Rate (WER)

Speech Recognition on CALLHOME En

Word Error Rate (WER)

Speech Recognition on CALLHOME Spanish Speech

WER

Speech Recognition on CAS-VSR-S101

Word Error Rate (WER)

Speech Recognition on GigaSpeech

Word Error Rate (WER)

Speech Recognition on Google Speech Commands - Musan

Error rate - SNR 0dB

Speech Recognition on Hub5'00 CallHome

Word Error Rate (WER)

Speech Recognition on Hub5'00 FISHER-SWBD

Word Error Rate (WER)

Speech Recognition on LRW

ESTOIPESQSTOI

Speech Recognition on LibriSpeech 100h test-clean

Word Error Rate (WER)

Speech Recognition on LibriSpeech 100h test-other

Word Error Rate (WER)

Speech Recognition on LibriSpeech train-clean-100 test-clean

Word Error Rate (WER)

Speech Recognition on LibriSpeech train-clean-100 test-other

Word Error Rate (WER)

Speech Recognition on Switchboard (300hr)

Word Error Rate (WER)

Speech Recognition on Switchboard CallHome

Word Error Rate (WER)

Speech Recognition on Switchboard SWBD

Word Error Rate (WER)

Speech Recognition on TCD-TIMIT corpus (mixed-speech)

ESTOIPESQSTOI

Speech Recognition on ToN_IoT

Accuray

Speech Recognition on VibraVox (forehead accelerometer)

Test PER

Speech Recognition on VibraVox (headset microphone)

Test PER

Speech Recognition on VibraVox (rigid in-ear microphone)

Test PER

Speech Recognition on VibraVox (soft in-ear microphone)

Test PER

Speech Recognition on VibraVox (temple vibration pickup)

Test PER

Speech Recognition on VibraVox (throat microphone)

Test PER

Speech Recognition on facebook/multilingual_librispeech german

WER

Speech Recognition on Common Voice

Test WER

Speech Recognition on Common Voice English

Word Error Rate (WER)Test WER

Speech Recognition on Common Voice French

Test WER

Speech Recognition on Common Voice Frisian

Test WER

Speech Recognition on Common Voice German

Test WERTest CER

Speech Recognition on Common Voice Italian

Test WER

Speech Recognition on Common Voice Japanese

Test WERTest CER

Speech Recognition on Common Voice Portuguese

Test WER

Speech Recognition on Common Voice Russian

Test WERTest CERTest CER (+LM)Test WER (+LM)

Speech Recognition on Common Voice Spanish

Test WERTest CERTest CER (+LM)Test WER (+LM)

Speech Recognition on HUI speech corpus

WER (%)

Speech Recognition on M-AILabs speech dataset

WER (%)

Speech Recognition on TUDA

Test WER

Speech Recognition on The Spoken Wikipedia Corpora

WER (%)

Speech Recognition on VoxPopuli

WER (%)

Speech Recognition on Voxforge German

WER (%)