Codesota · Speech · Speech Recognition · Earnings-22Tasks/Speech/Speech Recognition
Speech Recognition · benchmark dataset · 2022 · EN

Earnings-22 ASR Benchmark.

Earnings-22 is a 119-hour corpus of real-world English earnings calls covering 22 publicly traded companies. Heavy domain vocabulary (financial terminology, proper names) and accented speech make it a tough benchmark for production ASR systems.

Paper ↗Submit a result ↵
§ 01 · Leaderboard

Best published scores.

50 results indexed across 1 metric. Shaded row marks current SOTA; ties broken by submission date.

Primary
wer · lower is better
wer· primary
50 rows
#ModelOrgSubmittedPaper / codewer
01Granite Speech 4.1 2BOpenIBMMay 2025Granite-speech: open-source speech-aware LLMs with stron…8.37
02SYMPHONY-ASR—Jan 2026pwc-dump9.45
03Phi-4 Multimodal InstructOpenMicrosoftMar 2025Phi-4-Mini Technical Report: Compact yet Powerful Multim…10.16
04Granite Speech 3.3 2BOpenIBMMay 2025Granite-speech: open-source speech-aware LLMs with stron…10.25
05Canary-Qwen-2.5BOpenNVIDIAMar 2025Training and Inference Efficiency of Encoder-Decoder Spe…10.45
06Cohere Transcribe (Mar 2026)OpenCohereMar 2026pwc-dump10.86
07Parakeet-tdt-0.6b-v2—Apr 2023Efficient Sequence Transduction by Jointly Predicting To… · code11.15
08Lite-whisper-large-v3-turbo-acc—Feb 2025LiteASR: Efficient Automatic Speech Recognition with Low… · code11.53
09Whisper Large v3 TurboOpenOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code11.63
10Distil-large-v3—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code11.79
11Voxtral-Mini-4B-Realtime-2602OpenMistral AIFeb 2026Voxtral Realtime11.84
12Whisper Large v2OpenOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code12.05
13Niagara-38m-batch.en—Feb 2026pwc-dump12.09
14Distil-large-v2—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code12.19
15Parakeet-tdt_ctc-110m—Apr 2023Efficient Sequence Transduction by Jointly Predicting To… · code12.37
16Lite-whisper-large-v3-fast—Feb 2025LiteASR: Efficient Automatic Speech Recognition with Low… · code12.48
17Whisper-medium.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code12.63
18Whisper Large—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code12.91
19Whisper-small.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code12.97
20Distil-medium.en—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code12.99
21Distil-small.en—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code13.15
22VibeVoice-ASR-HF—Jan 2026VIBEVOICE-ASR Technical Report13.17
23Moonshine-streaming-small—Jan 2026pwc-dump13.55
24Owsm_ctc_v3.1_1B—Jan 2024OWSM v3.1: Better and Faster Open Whisper-Style Speech M… · code13.74
25Parakeet-ctc-0.6b—May 2023Fast Conformer with Linearly Scalable Attention for Effi…14.26
26Niagara-19m-batch.en—Feb 2026pwc-dump14.52
27Parakeet-rnnt-0.6b—May 2023Fast Conformer with Linearly Scalable Attention for Effi…14.66
28Asr-conformer-loquacious—Feb 2025pwc-dump14.71
29Whisper-base.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code15.09
30Stt_en_conformer_ctc_large—May 2020Conformer: Convolution-augmented Transformer for Speech … · code15.99
31Moonshine-base—Oct 2024Moonshine: Speech Recognition for Live Transcription and… · code16.85
32Stt_en_fastconformer_ctc_large—May 2023Fast Conformer with Linearly Scalable Attention for Effi…18.81
33Stt_en_conformer_ctc_small—May 2020Conformer: Convolution-augmented Transformer for Speech … · code18.84
34Whisper-tiny.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code19.12
35Stt_en_fastconformer_transducer_large—May 2023Fast Conformer with Linearly Scalable Attention for Effi…19.41
36Moonshine-streaming-tiny—Jan 2026pwc-dump20.19
37Moonshine-tiny—Oct 2024Moonshine: Speech Recognition for Live Transcription and… · code20.73
38Asr-wav2vec2-librispeech—Jun 2021SpeechBrain: A General-Purpose Speech Toolkit · code28.52
39Mms-1b-all—May 2023Scaling Speech Technology to 1,000+ Languages · code31.17
40Wav2vec2-large-960h-lv60-self—Jun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of… · code31.68
41Hubert-large-ls960-ft—Jun 2021HuBERT: Self-Supervised Speech Representation Learning b… · code35.24
42Hubert-xlarge-ls960-ft—Jun 2021HuBERT: Self-Supervised Speech Representation Learning b… · code36.13
43Wav2vec2-large-robust-ft-libri-960h—Apr 2021Robust wav2vec 2.0: Analyzing Domain Shift in Self-Super… · code36.22
44Wav2vec2-conformer-rope-large-960h-ft—Oct 2020fairseq S2T: Fast Speech-to-Text Modeling with fairseq · code37.52
45Data2vec-audio-large-960h—Feb 2022data2vec: A General Framework for Self-supervised Learni… · code37.82
46Wav2vec2-conformer-rel-pos-large-960h-ft—Oct 2020fairseq S2T: Fast Speech-to-Text Modeling with fairseq · code38.33
47wav2vec 2.0 Large (960h)OpenMeta AIJun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of… · code43.75
48Wav2vec2-base-960h—Jun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of… · code48.47
49Data2vec-audio-base-960h—Feb 2022data2vec: A General Framework for Self-supervised Learni… · code49.56
50Mms-1b-fl102—May 2023Scaling Speech Technology to 1,000+ Languages · code51.87
Fig 2 · Rows sorted by score within each metric. Shaded row marks SOTA. Dates reflect model or paper release where available, otherwise the date Codesota accessed the source.
§ 03 · Progress

5 steps
of state of the art.

Each row below marks a model that broke the previous record on wer. Intermediate submissions are kept in the leaderboard above; only SOTA-setting entries are re-listed here.

Lower scores win. Each subsequent entry improved upon the previous best.

SOTA line · wer
  1. May 16, 2020Stt_en_conformer_ctc_large15.99
  2. Dec 6, 2022Whisper Large v3 TurboOpenAI11.63
  3. Apr 13, 2023Parakeet-tdt-0.6b-v211.15
  4. Mar 3, 2025Phi-4 Multimodal InstructMicrosoft10.16
  5. May 13, 2025Granite Speech 4.1 2BIBM8.37
Fig 3 · SOTA-setting models only. 5 entries span May 2020 → May 2025.
§ 04 · Literature

20 papers
tied to this benchmark.

Every paper below corresponds to at least one row in the leaderboard above. Click through for the arXiv preprint and, when available, the reference implementation.

§ 06 · Contribute

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

Submit a result ↵Read submission guide
What a submission needs
  • 01A public checkpoint or API endpoint
  • 02A reproduction script with frozen commit + seed
  • 03Declared evaluation environment (Python, deps)
  • 04One row per metric declared by this dataset
  • 05A contact so we can follow up on discrepancies