TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Speech Recognition	AMI-IHM	Whisper	Word Error Rate (WER)	16.4	# 1
Speech Recognition	AMI SDM1	Whisper	Word Error Rate (WER)	36.9	# 2
Speech Recognition	Artie Bias Corpus	Whisper	WER	6.7	# 1
Speech Recognition	CALLHOME	Whisper	Word Error Rate (WER)	15.8	# 1
Speech Recognition	CHiME6	Whisper	WER	25.6	# 1
Speech Recognition	Common Voice	Whisper	Word Error Rate (WER)	9.5	# 1
Speech Recognition	CORAAL	Whisper	Word Error Rate (WER)	19.4	# 1
Spoken language identification	Fleurs	Whisper	Accuracy (%)	64.1	# 2
Speech Recognition	Fleurs (English)	Whisper	WER	4.6	# 1
Speech Recognition	LibriSpeech test-clean	Whisper	Word Error Rate (WER)	2.7	# 30
Speech Recognition	LibriSpeech test-other	Whisper	Word Error Rate (WER)	5.6	# 25
Speech Recognition	Switchboard corpus	Whisper	Word Error Rate (WER)	13.1	# 1
Speech Recognition	Tedlium	Whisper	Word Error Rate (WER)	4.0	# 1
Speech Recognition	Vox Populi	Whisper	Word Error Rate (WER)	7.3	# 1
Speech Recognition	WSJ	Whisper	Word Error Rate (WER)	3.1	# 1

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-ami-ihm)](https://paperswithcode.com/sota/speech-recognition-on-ami-ihm?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-artie-bias-corpus)](https://paperswithcode.com/sota/speech-recognition-on-artie-bias-corpus?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-callhome)](https://paperswithcode.com/sota/speech-recognition-on-callhome?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-chime6)](https://paperswithcode.com/sota/speech-recognition-on-chime6?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-common-voice-2)](https://paperswithcode.com/sota/speech-recognition-on-common-voice-2?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-coraal)](https://paperswithcode.com/sota/speech-recognition-on-coraal?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-fleurs-english)](https://paperswithcode.com/sota/speech-recognition-on-fleurs-english?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-switchboard-corpus)](https://paperswithcode.com/sota/speech-recognition-on-switchboard-corpus?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-tedlium)](https://paperswithcode.com/sota/speech-recognition-on-tedlium?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-vox-populi)](https://paperswithcode.com/sota/speech-recognition-on-vox-populi?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-wsj)](https://paperswithcode.com/sota/speech-recognition-on-wsj?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-ami-sdm1)](https://paperswithcode.com/sota/speech-recognition-on-ami-sdm1?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/spoken-language-identification-on-fleurs)](https://paperswithcode.com/sota/spoken-language-identification-on-fleurs?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-librispeech-test-other)](https://paperswithcode.com/sota/speech-recognition-on-librispeech-test-other?p=robust-speech-recognition-via-large-scale-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/robust-speech-recognition-via-large-scale-1/speech-recognition-on-librispeech-test-clean)](https://paperswithcode.com/sota/speech-recognition-on-librispeech-test-clean?p=robust-speech-recognition-via-large-scale-1)`

Robust Speech Recognition via Large-Scale Weak Supervision

Preprint 2022 · Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, Ilya Sutskever ·

We study the capabilities of speech processing systems trained simply to predict large amounts of transcripts of audio on the internet. When scaled to 680,000 hours of multilingual and multitask supervision, the resulting models generalize well to standard benchmarks and are often competitive with prior fully supervised results but in a zero-shot transfer setting without the need for any fine-tuning. When compared to humans, the models approach their accuracy and robustness. We are releasing models and inference code to serve as a foundation for further work on robust speech processing.

PDF Abstract Preprint 2022 PDF Preprint 2022 Abstract

Code

Add Remove Mark official

openai/whisper official

↳ Quickstart in

Colab

23,096

Tasks

Add Remove

Robust Speech Recognition

speech-recognition

Speech Recognition

Datasets

LibriSpeech

Common Voice VoxPopuli TED-LIUM 3 Multilingual LibriSpeech VoxLingua107 Switchboard-1 Corpus Fleurs Earnings-21 Artie Bias Corpus

Results from the Paper

Edit

Ranked #1 on Speech Recognition on CHiME6

Get a GitHub badge

Task	Dataset	Model	Metric Name	Metric Value	Global Rank	Benchmark
Speech Recognition	AMI-IHM	Whisper	Word Error Rate (WER)	16.4	# 1	Compare
Speech Recognition	AMI SDM1	Whisper	Word Error Rate (WER)	36.9	# 2	Compare
Speech Recognition	Artie Bias Corpus	Whisper	WER	6.7	# 1	Compare
Speech Recognition	CALLHOME	Whisper	Word Error Rate (WER)	15.8	# 1	Compare
Speech Recognition	CHiME6	Whisper	WER	25.6	# 1	Compare
Speech Recognition	Common Voice	Whisper	Word Error Rate (WER)	9.5	# 1	Compare
Speech Recognition	CORAAL	Whisper	Word Error Rate (WER)	19.4	# 1	Compare
Spoken language identification	Fleurs	Whisper	Accuracy (%)	64.1	# 2	Compare
Speech Recognition	Fleurs (English)	Whisper	WER	4.6	# 1	Compare
Speech Recognition	LibriSpeech test-clean	Whisper	Word Error Rate (WER)	2.7	# 30	Compare
Speech Recognition	LibriSpeech test-other	Whisper	Word Error Rate (WER)	5.6	# 25	Compare
Speech Recognition	Switchboard corpus	Whisper	Word Error Rate (WER)	13.1	# 1	Compare
Speech Recognition	Tedlium	Whisper	Word Error Rate (WER)	4.0	# 1	Compare
Speech Recognition	Vox Populi	Whisper	Word Error Rate (WER)	7.3	# 1	Compare
Speech Recognition	WSJ	Whisper	Word Error Rate (WER)	3.1	# 1	Compare

Methods

Add Remove

No methods listed for this paper. Add relevant methods here

Edit Social Preview

Robust Speech Recognition via Large-Scale Weak Supervision

Code Edit Add Remove Mark official

Tasks Edit Add Remove

Datasets Edit

Results from the Paper Edit

Methods Edit Add Remove

Code

Add Remove Mark official

Tasks

Add Remove

Datasets

Results from the Paper

Edit

Methods

Add Remove