Codesota · Speech · Speech Recognition · SPGISpeechTasks/Speech/Speech Recognition
Speech Recognition · benchmark dataset · 2021 · EN

SPGISpeech Earnings Call Corpus.

SPGISpeech is a 5,000-hour corpus of professionally transcribed English earnings calls released by S&P Global Market Intelligence. The largest publicly available financial-domain ASR benchmark.

Paper ↗Submit a result ↵
§ 01 · Leaderboard

Best published scores.

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

Primary
wer · lower is better
wer· primary
56 rows
#ModelOrgSubmittedPaper / codewer
01Audio Flamingo 3—Jul 2025Audio Flamingo 3: Advancing Audio Intelligence with Full… · code1.86
02Canary-Qwen-2.5BOpenNVIDIAMar 2025Training and Inference Efficiency of Encoder-Decoder Spe…1.90
03Canary-1B-FlashOpenNVIDIAMar 2025Training and Inference Efficiency of Encoder-Decoder Spe…1.95
04Parakeet-tdt-0.6b-v2—Apr 2023Efficient Sequence Transduction by Jointly Predicting To… · code2.17
05SYMPHONY-ASR—Jan 2026pwc-dump2.29
06Voxtral-Mini-3B-2507—Jul 2025Voxtral2.37
07Voxtral-Mini-4B-Realtime-2602OpenMistral AIFeb 2026Voxtral Realtime2.42
08Parakeet-tdt_ctc-110m—Apr 2023Efficient Sequence Transduction by Jointly Predicting To… · code2.54
09Qwen3-ASR-1.7BOpenAlibabaJan 2026Qwen3-ASR Technical Report · code2.84
10Distil-large-v3.5—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code2.87
11Owsm_ctc_v3.1_1B—Jan 2024OWSM v3.1: Better and Faster Open Whisper-Style Speech M… · code2.87
12Lite-whisper-large-v3-turbo-acc—Feb 2025LiteASR: Efficient Automatic Speech Recognition with Low… · code2.93
13Whisper Large v3OpenOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code2.94
14Whisper Large v3 TurboOpenOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code2.97
15Phi-4 Multimodal InstructOpenMicrosoftMar 2025Phi-4-Mini Technical Report: Compact yet Powerful Multim…3.06
16Cohere Transcribe (Mar 2026)OpenCohereMar 2026pwc-dump3.08
17Niagara-38m-batch.en—Feb 2026pwc-dump3.10
18Lite-whisper-large-v3-fast—Feb 2025LiteASR: Efficient Automatic Speech Recognition with Low… · code3.15
19Whisper Large—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code3.20
20Distil-large-v3—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code3.27
21Distil-large-v2—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code3.30
22Parakeet-rnnt-0.6b—May 2023Fast Conformer with Linearly Scalable Attention for Effi…3.32
23Whisper-medium.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code3.33
24Whisper-small.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code3.60
25Granite Speech 4.1 2BOpenIBMMay 2025Granite-speech: open-source speech-aware LLMs with stron…3.78
26VibeVoice-ASR-HF—Jan 2026VIBEVOICE-ASR Technical Report3.80
27Distil-small.en—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code3.82
28Distil-medium.en—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code3.83
29Whisper Large v2OpenOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code3.87
30Granite Speech 3.3 2BOpenIBMMay 2025Granite-speech: open-source speech-aware LLMs with stron…3.87
31Parakeet-ctc-0.6b—May 2023Fast Conformer with Linearly Scalable Attention for Effi…3.89
32Granite Speech 3.3 8BOpenIBMMay 2025Granite-speech: open-source speech-aware LLMs with stron…3.91
33Parakeet-tdt-0.6b-v3—Apr 2023Efficient Sequence Transduction by Jointly Predicting To… · code3.98
34Asr-conformer-loquacious—Feb 2025pwc-dump4.11
35Whisper-base.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code4.26
36Stt_en_fastconformer_transducer_large—May 2023Fast Conformer with Linearly Scalable Attention for Effi…4.97
37Stt_en_fastconformer_ctc_large—May 2023Fast Conformer with Linearly Scalable Attention for Effi…5.06
38Moonshine-base—Oct 2024Moonshine: Speech Recognition for Live Transcription and… · code5.46
39Stt_en_conformer_ctc_large—May 2020Conformer: Convolution-augmented Transformer for Speech … · code5.57
40Whisper-tiny.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code5.93
41Moonshine-streaming-tiny—Jan 2026pwc-dump6.16
42Moonshine-tiny—Oct 2024Moonshine: Speech Recognition for Live Transcription and… · code7.43
43Stt_en_conformer_ctc_small—May 2020Conformer: Convolution-augmented Transformer for Speech … · code7.80
44Asr-wav2vec2-librispeech—Jun 2021SpeechBrain: A General-Purpose Speech Toolkit · code10.39
45Mms-1b-all—May 2023Scaling Speech Technology to 1,000+ Languages · code16.95
46Wav2vec2-large-960h-lv60-self—Jun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of… · code17.94
47Data2vec-audio-large-960h—Feb 2022data2vec: A General Framework for Self-supervised Learni… · code18.49
48Hubert-xlarge-ls960-ft—Jun 2021HuBERT: Self-Supervised Speech Representation Learning b… · code18.58
49Wav2vec2-conformer-rel-pos-large-960h-ft—Oct 2020fairseq S2T: Fast Speech-to-Text Modeling with fairseq · code18.85
50Hubert-large-ls960-ft—Jun 2021HuBERT: Self-Supervised Speech Representation Learning b… · code18.86
51Wav2vec2-conformer-rope-large-960h-ft—Oct 2020fairseq S2T: Fast Speech-to-Text Modeling with fairseq · code18.87
52Wav2vec2-large-robust-ft-libri-960h—Apr 2021Robust wav2vec 2.0: Analyzing Domain Shift in Self-Super… · code19.03
53wav2vec 2.0 Large (960h)OpenMeta AIJun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of… · code22.82
54Data2vec-audio-base-960h—Feb 2022data2vec: A General Framework for Self-supervised Learni… · code25.46
55Mms-1b-fl102—May 2023Scaling Speech Technology to 1,000+ Languages · code26.21
56Wav2vec2-base-960h—Jun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of… · code27.56
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_large5.57
  2. Dec 6, 2022Whisper Large v3OpenAI2.94
  3. Apr 13, 2023Parakeet-tdt-0.6b-v22.17
  4. Mar 7, 2025Canary-Qwen-2.5BNVIDIA1.90
  5. Jul 10, 2025Audio Flamingo 31.86
Fig 3 · SOTA-setting models only. 5 entries span May 2020 → Jul 2025.
§ 04 · Literature

23 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