Dataset or benchmark

Open ASR Leaderboard: English short-form results (english_short_latest.csv)

hf-audio (Hugging Face), Hugging Face, hf-audio organization, 2026

Read the original

Read on October 19, 2026

What it is: The results file of the Open ASR Leaderboard, the Hugging Face Space that ranks speech recognition models. Each row is a model, with its average word error rate across the leaderboard's English short-form test sets, its inverse real-time factor, its license and its training data disclosure. The file was last modified on 17 October 2026.
How much weight it carries: Published by the leaderboard's own organization and listed among the datasets of the leaderboard Space. The average hides differences between test sets, so compare systems on the same test set. Several entries are vendor submissions, and the 'Proprietary' license marks closed systems that cannot be inspected. Results change as models are added.

Facts taken from this source

On the Open ASR Leaderboard's English short-form results, abr-ai/niagara-19m-batch.en had an average word error rate of 13.28 percent.
“abr-ai/niagara-19m-batch.en,13.27714285714286,4348.44” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, abr-ai/niagara-38m-batch.en had an average word error rate of 11.39 percent.
“abr-ai/niagara-38m-batch.en,11.388571428571426,4561.35” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, abr-ai/niagara-84m-batch.en had an average word error rate of 9.88 percent.
“abr-ai/niagara-84m-batch.en,9.881428571428572,1940.83” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, abr-ai/niagara-9m-batch.en had an average word error rate of 16.63 percent.
“abr-ai/niagara-9m-batch.en,16.634285714285713,7217.02” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, assemblyai/universal-3-5-pro had an average word error rate of 5.80 percent.
“assemblyai/universal-3-5-pro,5.8” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, bosonai/higgs-audio-v3-stt had an average word error rate of 6.19 percent.
“bosonai/higgs-audio-v3-stt,6.1899999999999995,109.65” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, CohereLabs/cohere-transcribe-03-2026 had an average word error rate of 6.26 percent.
“CohereLabs/cohere-transcribe-03-2026,6.2628571428571425,923.71” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, distil-whisper/distil-large-v3.5 had an average word error rate of 7.02 percent.
“distil-whisper/distil-large-v3.5,7.022857142857143,898.21” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, Edge0/ARK-ASR-0.6B had an average word error rate of 6.29 percent.
“Edge0/ARK-ASR-0.6B,6.291428571428571,673.02” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, Edge0/ARK-ASR-3B had an average word error rate of 6.53 percent.
“Edge0/ARK-ASR-3B,6.525714285714286,480.07” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, Edge0/Audio8-ASR-0.1B had an average word error rate of 8.75 percent.
“Edge0/Audio8-ASR-0.1B,8.754285714285714,725.77” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, elevenlabs/scribe_v2 had an average word error rate of 5.37 percent.
“elevenlabs/scribe_v2,5.365714285714285” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, espnet/owsm_ctc_v4_1B had an average word error rate of 8.55 percent.
“espnet/owsm_ctc_v4_1B,8.554285714285713,783.3” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, facebook/omniASR-CTC-7B-v2 had an average word error rate of 12.01 percent.
“facebook/omniASR-CTC-7B-v2,12.005714285714285,522.26” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, facebook/omniASR-LLM-7B-v2 had an average word error rate of 8.78 percent.
“facebook/omniASR-LLM-7B-v2,8.78,129.8” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, FermionResearch/Phonon-2 had an average word error rate of 7.23 percent.
“FermionResearch/Phonon-2,7.228571428571429,1976.52” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, FunAudioLLM/Fun-ASR-Nano-2512-hf had an average word error rate of 7.76 percent.
“FunAudioLLM/Fun-ASR-Nano-2512-hf,7.761428571428572,506.37” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, gladia/solaria-3 had an average word error rate of 18.90 percent.
“gladia/solaria-3,18.898571428571426” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, google/gemma-4-12B-it had an average word error rate of 20.87 percent.
“google/gemma-4-12B-it,20.86875,106.08” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, google/gemma-4-E2B-it had an average word error rate of 11.65 percent.
“google/gemma-4-E2B-it,11.64625,190.75” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, google/gemma-4-E4B-it had an average word error rate of 12.66 percent.
“google/gemma-4-E4B-it,12.66,160.69” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, HojoAI/Hojo-ASR-V1 had an average word error rate of 6.07 percent.
“HojoAI/Hojo-ASR-V1,6.065714285714286,73.39” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, ibm-granite/granite-4.0-1b-speech had an average word error rate of 7.02 percent.
“ibm-granite/granite-4.0-1b-speech,7.0200000000000005,666.99” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, ibm-granite/granite-speech-3.3-2b had an average word error rate of 5.36 percent.
“ibm-granite/granite-speech-3.3-2b,5.3575,509.35” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, ibm-granite/granite-speech-3.3-8b had an average word error rate of 5.26 percent.
“ibm-granite/granite-speech-3.3-8b,5.2625,262.79” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, ibm-granite/granite-speech-4.1-2b had an average word error rate of 6.42 percent.
“ibm-granite/granite-speech-4.1-2b,6.421428571428571,546.84” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, ibm-granite/granite-speech-4.1-2b-nar had an average word error rate of 6.45 percent.
“ibm-granite/granite-speech-4.1-2b-nar,6.45,2112.58” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, ibm-granite/granite-speech-5.0-470m-turboctc had an average word error rate of 6.84 percent.
“ibm-granite/granite-speech-5.0-470m-turboctc,6.837142857142857,21134.31” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, ibm-granite/granite-speech-5.0-470m-turboctc-nc had an average word error rate of 6.41 percent.
“ibm-granite/granite-speech-5.0-470m-turboctc-nc,6.407142857142857,20801.33” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, meta/muse-voice-transcribe had an average word error rate of 6.48 percent.
“meta/muse-voice-transcribe,6.482857142857143” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, microsoft/azure-speech-06-2026 had an average word error rate of 4.26 percent.
“microsoft/azure-speech-06-2026,4.25875” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, microsoft/azure-speech-07-2026 had an average word error rate of 5.21 percent.
“microsoft/azure-speech-07-2026,5.207142857142856” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, microsoft/Phi-4-multimodal-instruct had an average word error rate of 5.02 percent.
“microsoft/Phi-4-multimodal-instruct,5.02125,162.53” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, microsoft/VibeVoice-ASR-HF had an average word error rate of 8.23 percent.
“microsoft/VibeVoice-ASR-HF,8.227142857142857,217.89” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, mistralai/Voxtral-Mini-3B-2507 had an average word error rate of 7.84 percent.
“mistralai/Voxtral-Mini-3B-2507,7.837142857142856,179.78” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, mistralai/Voxtral-Mini-4B-Realtime-2602 had an average word error rate of 9.01 percent.
“mistralai/Voxtral-Mini-4B-Realtime-2602,9.012857142857142,103.48” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, mistralai/Voxtral-Small-24B-2507 had an average word error rate of 6.90 percent.
“mistralai/Voxtral-Small-24B-2507,6.895714285714285,101.45” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, modulate/multilingual had an average word error rate of 5.45 percent.
“modulate/multilingual,5.447142857142857” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, nvidia/canary-1b-v2 had an average word error rate of 7.91 percent.
“nvidia/canary-1b-v2,7.911428571428572,1853.68” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, nvidia/canary-qwen-2.5b had an average word error rate of 6.37 percent.
“nvidia/canary-qwen-2.5b,6.368571428571429,881.67” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, nvidia/nemotron-3.5-asr-streaming-0.6b had an average word error rate of 10.51 percent.
“nvidia/nemotron-3.5-asr-streaming-0.6b,10.505714285714287,1348.36” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, nvidia/nemotron-speech-streaming-en-0.6b had an average word error rate of 6.89 percent.
“nvidia/nemotron-speech-streaming-en-0.6b,6.894285714285714,1114.14” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, nvidia/parakeet-ctc-1.1b had an average word error rate of 8.22 percent.
“nvidia/parakeet-ctc-1.1b,8.215714285714288,5137.41” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, nvidia/parakeet-tdt-0.6b-v2 had an average word error rate of 6.43 percent.
“nvidia/parakeet-tdt-0.6b-v2,6.432857142857143,6148.32” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, nvidia/parakeet-tdt-0.6b-v2 (fast-gpu-asr) had an average word error rate of 6.44 percent.
“nvidia/parakeet-tdt-0.6b-v2 (fast-gpu-asr),6.437142857142858,14383.48” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, nvidia/parakeet-tdt-0.6b-v3 had an average word error rate of 6.51 percent.
“nvidia/parakeet-tdt-0.6b-v3,6.514285714285713,6227.73” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, nvidia/parakeet-tdt-0.6b-v3 (fast-gpu-asr) had an average word error rate of 6.38 percent.
“nvidia/parakeet-tdt-0.6b-v3 (fast-gpu-asr),6.377142857142857,13304.58” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, openai/gpt-transcribe had an average word error rate of 6.38 percent.
“openai/gpt-transcribe,6.384285714285714” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, openai/whisper-large-v3 had an average word error rate of 7.84 percent.
“openai/whisper-large-v3,7.841428571428572,470.73” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, openai/whisper-large-v3-turbo had an average word error rate of 8.39 percent.
“openai/whisper-large-v3-turbo,8.387142857142857,810.5” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, OpenMOSS-Team/MOSS-Transcribe-Diarize had an average word error rate of 6.29 percent.
“OpenMOSS-Team/MOSS-Transcribe-Diarize,6.2871428571428565,387.46” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, OtterAI/speech-v3.1 had an average word error rate of 3.88 percent.
“OtterAI/speech-v3.1,3.8844444444444446” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, Qwen/Qwen3-ASR-0.6B-hf had an average word error rate of 6.60 percent.
“Qwen/Qwen3-ASR-0.6B-hf,6.602857142857144,749.52” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, Qwen/Qwen3-ASR-1.7B-hf had an average word error rate of 5.75 percent.
“Qwen/Qwen3-ASR-1.7B-hf,5.745714285714286,835.62” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, reson8/resonant-1 had an average word error rate of 5.53 percent.
“reson8/resonant-1,5.531428571428571” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, reson8/resonant-1-flash had an average word error rate of 4.02 percent.
“reson8/resonant-1-flash,4.02375” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, smallestai/pulse had an average word error rate of 5.96 percent.
“smallestai/pulse,5.964285714285715” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, sophea/asr-k1 (preview) had an average word error rate of 4.61 percent.
“sophea/asr-k1 (preview),4.614444444444445” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, soundsgoodai/Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, ctc_greedy_search) had an average word error rate of 9.02 percent.
“ctc_greedy_search)",9.022857142857143,20886.37” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, soundsgoodai/Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search) had an average word error rate of 7.38 percent.
“transducer_modified_beam_search)",7.381428571428572,19289.67” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, sprag/symphony had an average word error rate of 6.33 percent.
“sprag/symphony,6.325714285714285” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, TheStageAI/thewhisper-large-v3-turbo had an average word error rate of 6.07 percent.
“TheStageAI/thewhisper-large-v3-turbo,6.074285714285714,1183.92” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, zoom/scribe_v1 had an average word error rate of 4.15 percent.
“zoom/scribe_v1,4.15” (english_short_latest.csv, column avg)
On the Open ASR Leaderboard's English short-form results, zoom/scribe_v2_pro had an average word error rate of 4.77 percent.
“zoom/scribe_v2_pro,4.768571428571428” (english_short_latest.csv, column avg)

Pages that rest on this source

Speech Recognition Accuracy: Word Error RatesDataset