{"dataset":{"name":"Word error rates of speech recognition models and APIs on the Open ASR Leaderboard's public English test sets (tracked weekly)","title":"Open ASR Leaderboard: Speech Recognition Accuracy, Tracked","question":"Which speech recognition models and APIs make the fewest errors on the same English tests, and how do open models compare with paid APIs?","url":"https://yapapp.cc/data/open-asr-leaderboard","measure":{"label":"Average word error rate (public test sets)","unit":"percent","short":"%","decimals":2,"lowerIsBetter":true},"dimensions":[{"key":"model","label":"Model"},{"key":"maker","label":"Maker"},{"key":"access","label":"Access","values":["Open weights","Cloud API"]},{"key":"product","label":"Product"},{"key":"license","label":"License"},{"key":"sizeB","label":"Size (billion parameters)"},{"key":"rtfx","label":"Speed (RTFx)"},{"key":"lsClean","label":"LibriSpeech clean"},{"key":"lsOther","label":"LibriSpeech other"},{"key":"ami","label":"AMI meetings"},{"key":"earnings22","label":"Earnings-22 calls"},{"key":"gigaspeech","label":"GigaSpeech"},{"key":"spgispeech","label":"SPGISpeech"},{"key":"voxpopuli","label":"VoxPopuli"},{"key":"urgent2024","label":"Noisy speech (URGENT 2024)"},{"key":"monsoon","label":"Indian English conversations"},{"key":"version","label":"Leaderboard version"}],"include":"Every model the Open ASR Leaderboard ranks in its English short-form table on the day read (scored on each of its public default test sets): open-weight models and cloud APIs, including the separate rows it keeps for one model run with a faster setup.","exclude":"Models the leaderboard leaves unranked because a default test set is missing, and its long-form, multilingual and private-data tables.","method":"Read once a week. We read the leaderboard's own configuration on Hugging Face (its latest version, the exact revision of the English results file it shows, and its default test sets), fetch that file at that revision, and record each model's word error rate on every public default test set and their mean, rounded to two decimals as the leaderboard computes it. Speed (RTFx), license and size are the leaderboard's columns as published; every reading is kept.","limitations":"The leaderboard's own Average WER also folds in two averages over private recordings that are not published, so its order can differ slightly from this public-set average. The test sets are audiobooks, meetings, earnings calls, podcasts, parliament speeches, noisy speech and conversations, not dictation, and text is normalized before scoring, so capitals and punctuation errors are not counted (our own formatting test covers those). The leaderboard adds sets and changes its defaults from time to time, which moves every average. Speed is measured on the leaderboard's own hardware; cloud APIs have none.","sources":0,"updated":"2026-10-20","license":"https://creativecommons.org/licenses/by/4.0/","credit":"Yap (2026). Word error rates of speech recognition models and APIs on the Open ASR Leaderboard's public English test sets (tracked weekly) [Data set]. https://yapapp.cc/data/open-asr-leaderboard"},"rows":[{"row":"Qwen/Qwen3-ASR-1.7B-hf","value":4.94,"low":null,"high":null,"as_written":"4.94","unit":"percent","model":"Qwen/Qwen3-ASR-1.7B-hf","maker":"Qwen","access":"Open weights","product":"Qwen3-ASR","license":"apache-2.0","sizeB":2.04,"rtfx":835.62,"lsClean":1.26,"lsOther":2.94,"ami":8.31,"earnings22":5.84,"gigaspeech":7.22,"spgispeech":2.57,"voxpopuli":2.85,"urgent2024":9.93,"monsoon":3.5,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), Qwen/Qwen3-ASR-1.7B-hf had an average word error rate of 4.94 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-qwen-qwen3-asr-1-7b-hf","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"CohereLabs/cohere-transcribe-03-2026","value":5.21,"low":null,"high":null,"as_written":"5.21","unit":"percent","model":"CohereLabs/cohere-transcribe-03-2026","maker":"CohereLabs","access":"Open weights","product":"Cohere Transcribe","license":"apache-2.0","sizeB":2,"rtfx":923.71,"lsClean":0.97,"lsOther":2.05,"ami":7.02,"earnings22":7.88,"gigaspeech":7.88,"spgispeech":2.74,"voxpopuli":5.38,"urgent2024":9.5,"monsoon":3.44,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), CohereLabs/cohere-transcribe-03-2026 had an average word error rate of 5.21 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-coherelabs-cohere-transcribe-03-2026","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"bosonai/higgs-audio-v3-stt","value":5.22,"low":null,"high":null,"as_written":"5.22","unit":"percent","model":"bosonai/higgs-audio-v3-stt","maker":"bosonai","access":"Open weights","product":null,"license":"apache-2.0","sizeB":2.68,"rtfx":109.65,"lsClean":1.08,"lsOther":2.61,"ami":7.19,"earnings22":7.04,"gigaspeech":7.45,"spgispeech":2.07,"voxpopuli":3.73,"urgent2024":11.88,"monsoon":3.97,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), bosonai/higgs-audio-v3-stt had an average word error rate of 5.22 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-bosonai-higgs-audio-v3-stt","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"HojoAI/Hojo-ASR-V1","value":5.22,"low":null,"high":null,"as_written":"5.22","unit":"percent","model":"HojoAI/Hojo-ASR-V1","maker":"HojoAI","access":"Open weights","product":null,"license":"apache-2.0","sizeB":5.18,"rtfx":73.39,"lsClean":1.3,"lsOther":3.24,"ami":7.54,"earnings22":6.97,"gigaspeech":6.14,"spgispeech":1.74,"voxpopuli":3.11,"urgent2024":12.33,"monsoon":4.63,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), HojoAI/Hojo-ASR-V1 had an average word error rate of 5.22 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-hojoai-hojo-asr-v1","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"TheStageAI/thewhisper-large-v3-turbo","value":5.24,"low":null,"high":null,"as_written":"5.24","unit":"percent","model":"TheStageAI/thewhisper-large-v3-turbo","maker":"TheStageAI","access":"Open weights","product":null,"license":"cc-by-4.0","sizeB":0.8,"rtfx":1183.92,"lsClean":1.41,"lsOther":3.2,"ami":7.32,"earnings22":7.18,"gigaspeech":7.84,"spgispeech":1.62,"voxpopuli":4.02,"urgent2024":10.82,"monsoon":3.72,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), TheStageAI/thewhisper-large-v3-turbo had an average word error rate of 5.24 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-thestageai-thewhisper-large-v3-turbo","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"ibm-granite/granite-speech-4.1-2b","value":5.35,"low":null,"high":null,"as_written":"5.35","unit":"percent","model":"ibm-granite/granite-speech-4.1-2b","maker":"ibm-granite","access":"Open weights","product":"IBM Granite Speech","license":"apache-2.0","sizeB":2,"rtfx":546.84,"lsClean":1.03,"lsOther":2.19,"ami":7.06,"earnings22":6.69,"gigaspeech":8.18,"spgispeech":3.47,"voxpopuli":4.18,"urgent2024":11.25,"monsoon":4.12,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), ibm-granite/granite-speech-4.1-2b had an average word error rate of 5.35 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-ibm-granite-granite-speech-4-1-2b","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"nvidia/canary-qwen-2.5b","value":5.38,"low":null,"high":null,"as_written":"5.38","unit":"percent","model":"nvidia/canary-qwen-2.5b","maker":"nvidia","access":"Open weights","product":"NVIDIA Canary","license":"cc-by-4.0","sizeB":2.5,"rtfx":881.67,"lsClean":1.23,"lsOther":2.63,"ami":7.91,"earnings22":6.41,"gigaspeech":7.8,"spgispeech":1.7,"voxpopuli":4.14,"urgent2024":13.02,"monsoon":3.6,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), nvidia/canary-qwen-2.5b had an average word error rate of 5.38 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-nvidia-canary-qwen-2-5b","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"ibm-granite/granite-speech-4.1-2b-nar","value":5.4,"low":null,"high":null,"as_written":"5.4","unit":"percent","model":"ibm-granite/granite-speech-4.1-2b-nar","maker":"ibm-granite","access":"Open weights","product":"IBM Granite Speech","license":"apache-2.0","sizeB":2,"rtfx":2112.58,"lsClean":1.04,"lsOther":2.4,"ami":6.96,"earnings22":6.67,"gigaspeech":8.61,"spgispeech":3.23,"voxpopuli":4.25,"urgent2024":11.19,"monsoon":4.24,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), ibm-granite/granite-speech-4.1-2b-nar had an average word error rate of 5.40 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-ibm-granite-granite-speech-4-1-2b-nar","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"Edge0/ARK-ASR-0.6B","value":5.44,"low":null,"high":null,"as_written":"5.44","unit":"percent","model":"Edge0/ARK-ASR-0.6B","maker":"Edge0","access":"Open weights","product":null,"license":"apache-2.0","sizeB":1.15,"rtfx":673.02,"lsClean":1.48,"lsOther":3.44,"ami":8.61,"earnings22":6.75,"gigaspeech":7.37,"spgispeech":2.63,"voxpopuli":2.76,"urgent2024":12.49,"monsoon":3.43,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), Edge0/ARK-ASR-0.6B had an average word error rate of 5.44 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-edge0-ark-asr-0-6b","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"ibm-granite/granite-speech-5.0-470m-turboctc-nc","value":5.44,"low":null,"high":null,"as_written":"5.44","unit":"percent","model":"ibm-granite/granite-speech-5.0-470m-turboctc-nc","maker":"ibm-granite","access":"Open weights","product":"IBM Granite Speech","license":"cc-by-nc-sa-4.0","sizeB":0.47,"rtfx":20801.33,"lsClean":1.45,"lsOther":2.67,"ami":7.31,"earnings22":8.23,"gigaspeech":8.37,"spgispeech":1.75,"voxpopuli":4.67,"urgent2024":10.22,"monsoon":4.3,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), ibm-granite/granite-speech-5.0-470m-turboctc-nc had an average word error rate of 5.44 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-ibm-granite-granite-speech-5-0-470m-turboctc-nc","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"nvidia/parakeet-tdt-0.6b-v2 (fast-gpu-asr)","value":5.44,"low":null,"high":null,"as_written":"5.44","unit":"percent","model":"nvidia/parakeet-tdt-0.6b-v2 (fast-gpu-asr)","maker":"nvidia","access":"Open weights","product":"NVIDIA Parakeet","license":"cc-by-4.0","sizeB":0.6,"rtfx":14383.48,"lsClean":1.27,"lsOther":2.65,"ami":9.02,"earnings22":6.83,"gigaspeech":8.17,"spgispeech":1.87,"voxpopuli":3.89,"urgent2024":11.58,"monsoon":3.7,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), nvidia/parakeet-tdt-0.6b-v2 (fast-gpu-asr) had an average word error rate of 5.44 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-nvidia-parakeet-tdt-0-6b-v2-fast-gpu-asr","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"Edge0/ARK-ASR-3B","value":5.45,"low":null,"high":null,"as_written":"5.45","unit":"percent","model":"Edge0/ARK-ASR-3B","maker":"Edge0","access":"Open weights","product":null,"license":"apache-2.0","sizeB":3.75,"rtfx":480.07,"lsClean":1.04,"lsOther":2.37,"ami":7.89,"earnings22":8.07,"gigaspeech":6.94,"spgispeech":2.46,"voxpopuli":3.55,"urgent2024":13.29,"monsoon":3.48,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), Edge0/ARK-ASR-3B had an average word error rate of 5.45 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-edge0-ark-asr-3b","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"nvidia/parakeet-tdt-0.6b-v2","value":5.45,"low":null,"high":null,"as_written":"5.45","unit":"percent","model":"nvidia/parakeet-tdt-0.6b-v2","maker":"nvidia","access":"Open weights","product":"NVIDIA Parakeet","license":"cc-by-4.0","sizeB":0.6,"rtfx":6148.32,"lsClean":1.28,"lsOther":2.73,"ami":9.09,"earnings22":6.76,"gigaspeech":8.15,"spgispeech":1.94,"voxpopuli":3.78,"urgent2024":11.42,"monsoon":3.89,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), nvidia/parakeet-tdt-0.6b-v2 had an average word error rate of 5.45 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-nvidia-parakeet-tdt-0-6b-v2","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"nvidia/parakeet-tdt-0.6b-v3 (fast-gpu-asr)","value":5.46,"low":null,"high":null,"as_written":"5.46","unit":"percent","model":"nvidia/parakeet-tdt-0.6b-v3 (fast-gpu-asr)","maker":"nvidia","access":"Open weights","product":"NVIDIA Parakeet","license":"cc-by-4.0","sizeB":0.6,"rtfx":13304.58,"lsClean":1.46,"lsOther":3.08,"ami":9.02,"earnings22":5.81,"gigaspeech":7.75,"spgispeech":3.56,"voxpopuli":3.04,"urgent2024":11.5,"monsoon":3.96,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), nvidia/parakeet-tdt-0.6b-v3 (fast-gpu-asr) had an average word error rate of 5.46 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-nvidia-parakeet-tdt-0-6b-v3-fast-gpu-asr","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"OpenMOSS-Team/MOSS-Transcribe-Diarize","value":5.54,"low":null,"high":null,"as_written":"5.54","unit":"percent","model":"OpenMOSS-Team/MOSS-Transcribe-Diarize","maker":"OpenMOSS-Team","access":"Open weights","product":null,"license":"apache-2.0","sizeB":0.91,"rtfx":387.46,"lsClean":1.66,"lsOther":4.16,"ami":8.33,"earnings22":6.09,"gigaspeech":7.68,"spgispeech":1.7,"voxpopuli":3.6,"urgent2024":12.74,"monsoon":3.87,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), OpenMOSS-Team/MOSS-Transcribe-Diarize had an average word error rate of 5.54 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-openmoss-team-moss-transcribe-diarize","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"nvidia/parakeet-tdt-0.6b-v3","value":5.58,"low":null,"high":null,"as_written":"5.58","unit":"percent","model":"nvidia/parakeet-tdt-0.6b-v3","maker":"nvidia","access":"Open weights","product":"NVIDIA Parakeet","license":"cc-by-4.0","sizeB":0.6,"rtfx":6227.73,"lsClean":1.52,"lsOther":3.13,"ami":9.42,"earnings22":5.85,"gigaspeech":7.99,"spgispeech":3.63,"voxpopuli":3.05,"urgent2024":11.52,"monsoon":4.14,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), nvidia/parakeet-tdt-0.6b-v3 had an average word error rate of 5.58 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-nvidia-parakeet-tdt-0-6b-v3","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"Qwen/Qwen3-ASR-0.6B-hf","value":5.77,"low":null,"high":null,"as_written":"5.77","unit":"percent","model":"Qwen/Qwen3-ASR-0.6B-hf","maker":"Qwen","access":"Open weights","product":"Qwen3-ASR","license":"apache-2.0","sizeB":0.78,"rtfx":749.52,"lsClean":1.7,"lsOther":4.01,"ami":9.33,"earnings22":7.83,"gigaspeech":7.62,"spgispeech":2.74,"voxpopuli":2.9,"urgent2024":11.57,"monsoon":4.23,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), Qwen/Qwen3-ASR-0.6B-hf had an average word error rate of 5.77 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-qwen-qwen3-asr-0-6b-hf","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"ibm-granite/granite-speech-5.0-470m-turboctc","value":5.78,"low":null,"high":null,"as_written":"5.78","unit":"percent","model":"ibm-granite/granite-speech-5.0-470m-turboctc","maker":"ibm-granite","access":"Open weights","product":"IBM Granite Speech","license":"apache-2.0","sizeB":0.47,"rtfx":21134.31,"lsClean":1.48,"lsOther":2.71,"ami":7.71,"earnings22":6.75,"gigaspeech":8.74,"spgispeech":3.22,"voxpopuli":4.9,"urgent2024":11.79,"monsoon":4.75,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), ibm-granite/granite-speech-5.0-470m-turboctc had an average word error rate of 5.78 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-ibm-granite-granite-speech-5-0-470m-turboctc","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"mistralai/Voxtral-Small-24B-2507","value":5.81,"low":null,"high":null,"as_written":"5.81","unit":"percent","model":"mistralai/Voxtral-Small-24B-2507","maker":"mistralai","access":"Open weights","product":"Voxtral","license":"apache-2.0","sizeB":24,"rtfx":101.45,"lsClean":1.23,"lsOther":2.8,"ami":13.19,"earnings22":7.07,"gigaspeech":8.33,"spgispeech":1.84,"voxpopuli":2,"urgent2024":12.35,"monsoon":3.49,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), mistralai/Voxtral-Small-24B-2507 had an average word error rate of 5.81 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-mistralai-voxtral-small-24b-2507","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"ibm-granite/granite-4.0-1b-speech","value":5.86,"low":null,"high":null,"as_written":"5.86","unit":"percent","model":"ibm-granite/granite-4.0-1b-speech","maker":"ibm-granite","access":"Open weights","product":"IBM Granite Speech","license":"apache-2.0","sizeB":2,"rtfx":666.99,"lsClean":1.11,"lsOther":2.5,"ami":7.38,"earnings22":8.53,"gigaspeech":8.44,"spgispeech":3.55,"voxpopuli":4.41,"urgent2024":11.99,"monsoon":4.84,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), ibm-granite/granite-4.0-1b-speech had an average word error rate of 5.86 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-ibm-granite-granite-4-0-1b-speech","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"nvidia/nemotron-speech-streaming-en-0.6b","value":6.06,"low":null,"high":null,"as_written":"6.06","unit":"percent","model":"nvidia/nemotron-speech-streaming-en-0.6b","maker":"nvidia","access":"Open weights","product":"NVIDIA Nemotron Speech ASR","license":"nvidia-open-model-license","sizeB":0.62,"rtfx":1114.14,"lsClean":1.89,"lsOther":4.35,"ami":8.84,"earnings22":9.45,"gigaspeech":7.9,"spgispeech":2.69,"voxpopuli":2.4,"urgent2024":12.54,"monsoon":4.44,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), nvidia/nemotron-speech-streaming-en-0.6b had an average word error rate of 6.06 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-nvidia-nemotron-speech-streaming-en-0-6b","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"distil-whisper/distil-large-v3.5","value":6.18,"low":null,"high":null,"as_written":"6.18","unit":"percent","model":"distil-whisper/distil-large-v3.5","maker":"distil-whisper","access":"Open weights","product":"Distil-Whisper","license":"mit","sizeB":0.8,"rtfx":898.21,"lsClean":1.94,"lsOther":4.52,"ami":12.09,"earnings22":7.7,"gigaspeech":8.21,"spgispeech":2.62,"voxpopuli":2.52,"urgent2024":12.42,"monsoon":3.6,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), distil-whisper/distil-large-v3.5 had an average word error rate of 6.18 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-distil-whisper-distil-large-v3-5","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"soundsgoodai/Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-a","value":6.22,"low":null,"high":null,"as_written":"6.22","unit":"percent","model":"soundsgoodai/Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search)","maker":"soundsgoodai","access":"Open weights","product":null,"license":"cc-by-nc-4.0","sizeB":0.29,"rtfx":19289.67,"lsClean":1.31,"lsOther":3.03,"ami":10.24,"earnings22":7.9,"gigaspeech":8.3,"spgispeech":1.64,"voxpopuli":4.31,"urgent2024":13.7,"monsoon":5.58,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), soundsgoodai/Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search) had an average word error rate of 6.22 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-soundsgoodai-zipformer-cr-ctc-transducer-xl-290m-fast-gpu-asr-transducer","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"FermionResearch/Phonon-2","value":6.25,"low":null,"high":null,"as_written":"6.25","unit":"percent","model":"FermionResearch/Phonon-2","maker":"FermionResearch","access":"Open weights","product":null,"license":"cc-by-4.0","sizeB":0.6,"rtfx":1976.52,"lsClean":1.73,"lsOther":3.92,"ami":9.32,"earnings22":6.96,"gigaspeech":8.35,"spgispeech":3.7,"voxpopuli":2.46,"urgent2024":14.9,"monsoon":4.91,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), FermionResearch/Phonon-2 had an average word error rate of 6.25 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-fermionresearch-phonon-2","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"FunAudioLLM/Fun-ASR-Nano-2512-hf","value":6.65,"low":null,"high":null,"as_written":"6.65","unit":"percent","model":"FunAudioLLM/Fun-ASR-Nano-2512-hf","maker":"FunAudioLLM","access":"Open weights","product":"FunASR","license":"apache-2.0","sizeB":0.83,"rtfx":506.37,"lsClean":1.47,"lsOther":4.07,"ami":10.35,"earnings22":10.87,"gigaspeech":8.41,"spgispeech":2.05,"voxpopuli":3.4,"urgent2024":13.93,"monsoon":5.32,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), FunAudioLLM/Fun-ASR-Nano-2512-hf had an average word error rate of 6.65 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-funaudiollm-fun-asr-nano-2512-hf","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"mistralai/Voxtral-Mini-3B-2507","value":6.66,"low":null,"high":null,"as_written":"6.66","unit":"percent","model":"mistralai/Voxtral-Mini-3B-2507","maker":"mistralai","access":"Open weights","product":"Voxtral","license":"apache-2.0","sizeB":5,"rtfx":179.78,"lsClean":1.48,"lsOther":3.63,"ami":13.57,"earnings22":7.66,"gigaspeech":8.7,"spgispeech":2.15,"voxpopuli":2.22,"urgent2024":15.66,"monsoon":4.9,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), mistralai/Voxtral-Mini-3B-2507 had an average word error rate of 6.66 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-mistralai-voxtral-mini-3b-2507","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"openai/whisper-large-v3","value":6.66,"low":null,"high":null,"as_written":"6.66","unit":"percent","model":"openai/whisper-large-v3","maker":"openai","access":"Open weights","product":"Whisper","license":"apache-2.0","sizeB":2,"rtfx":470.73,"lsClean":1.56,"lsOther":3.52,"ami":13.63,"earnings22":7.99,"gigaspeech":8.35,"spgispeech":2.71,"voxpopuli":4.53,"urgent2024":13.73,"monsoon":3.95,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), openai/whisper-large-v3 had an average word error rate of 6.66 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-openai-whisper-large-v3","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"nvidia/canary-1b-v2","value":6.69,"low":null,"high":null,"as_written":"6.69","unit":"percent","model":"nvidia/canary-1b-v2","maker":"nvidia","access":"Open weights","product":"NVIDIA Canary","license":"cc-by-4.0","sizeB":1,"rtfx":1853.68,"lsClean":1.76,"lsOther":3.07,"ami":13.03,"earnings22":7.28,"gigaspeech":9.12,"spgispeech":2.03,"voxpopuli":4.39,"urgent2024":14.56,"monsoon":4.97,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), nvidia/canary-1b-v2 had an average word error rate of 6.69 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-nvidia-canary-1b-v2","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"nvidia/parakeet-ctc-1.1b","value":6.92,"low":null,"high":null,"as_written":"6.92","unit":"percent","model":"nvidia/parakeet-ctc-1.1b","maker":"nvidia","access":"Open weights","product":"NVIDIA Parakeet","license":"cc-by-4.0","sizeB":1.1,"rtfx":5137.41,"lsClean":1.52,"lsOther":3.23,"ami":12.23,"earnings22":9.66,"gigaspeech":8.74,"spgispeech":3.53,"voxpopuli":3.24,"urgent2024":14.88,"monsoon":5.23,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), nvidia/parakeet-ctc-1.1b had an average word error rate of 6.92 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-nvidia-parakeet-ctc-1-1b","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"microsoft/VibeVoice-ASR-HF","value":7.12,"low":null,"high":null,"as_written":"7.12","unit":"percent","model":"microsoft/VibeVoice-ASR-HF","maker":"microsoft","access":"Open weights","product":"VibeVoice-ASR","license":"mit","sizeB":8,"rtfx":217.89,"lsClean":1.56,"lsOther":4.95,"ami":12.02,"earnings22":7.97,"gigaspeech":7.87,"spgispeech":3.47,"voxpopuli":2.02,"urgent2024":19.5,"monsoon":4.74,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), microsoft/VibeVoice-ASR-HF had an average word error rate of 7.12 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-microsoft-vibevoice-asr-hf","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"openai/whisper-large-v3-turbo","value":7.17,"low":null,"high":null,"as_written":"7.17","unit":"percent","model":"openai/whisper-large-v3-turbo","maker":"openai","access":"Open weights","product":"Whisper","license":"mit","sizeB":0.8,"rtfx":810.5,"lsClean":2.13,"lsOther":3.71,"ami":13.88,"earnings22":8.09,"gigaspeech":8.47,"spgispeech":2.79,"voxpopuli":7.02,"urgent2024":13.69,"monsoon":4.77,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), openai/whisper-large-v3-turbo had an average word error rate of 7.17 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-openai-whisper-large-v3-turbo","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"facebook/omniASR-LLM-7B-v2","value":7.34,"low":null,"high":null,"as_written":"7.34","unit":"percent","model":"facebook/omniASR-LLM-7B-v2","maker":"facebook","access":"Open weights","product":"Omnilingual ASR","license":"apache-2.0","sizeB":7.8,"rtfx":129.8,"lsClean":1.35,"lsOther":3.25,"ami":13.86,"earnings22":11.44,"gigaspeech":9.2,"spgispeech":3.22,"voxpopuli":4.03,"urgent2024":14.84,"monsoon":4.87,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), facebook/omniASR-LLM-7B-v2 had an average word error rate of 7.34 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-facebook-omniasr-llm-7b-v2","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"espnet/owsm_ctc_v4_1B","value":7.37,"low":null,"high":null,"as_written":"7.37","unit":"percent","model":"espnet/owsm_ctc_v4_1B","maker":"espnet","access":"Open weights","product":"OWSM","license":"cc-by-4.0","sizeB":1,"rtfx":783.3,"lsClean":2.11,"lsOther":4.38,"ami":11.1,"earnings22":10.31,"gigaspeech":9.29,"spgispeech":2.31,"voxpopuli":4.09,"urgent2024":16.12,"monsoon":6.66,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), espnet/owsm_ctc_v4_1B had an average word error rate of 7.37 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-espnet-owsm-ctc-v4-1b","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"soundsgoodai/Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-a","value":7.64,"low":null,"high":null,"as_written":"7.64","unit":"percent","model":"soundsgoodai/Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, ctc_greedy_search)","maker":"soundsgoodai","access":"Open weights","product":null,"license":"cc-by-nc-4.0","sizeB":0.29,"rtfx":20886.37,"lsClean":1.71,"lsOther":3.89,"ami":11.64,"earnings22":9.26,"gigaspeech":9.52,"spgispeech":2.75,"voxpopuli":4.53,"urgent2024":17.69,"monsoon":7.77,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), soundsgoodai/Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, ctc_greedy_search) had an average word error rate of 7.64 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-soundsgoodai-zipformer-cr-ctc-transducer-xl-290m-fast-gpu-asr-ctc-greedy","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"mistralai/Voxtral-Mini-4B-Realtime-2602","value":7.74,"low":null,"high":null,"as_written":"7.74","unit":"percent","model":"mistralai/Voxtral-Mini-4B-Realtime-2602","maker":"mistralai","access":"Open weights","product":"Voxtral","license":"apache-2.0","sizeB":4,"rtfx":103.48,"lsClean":1.62,"lsOther":4.94,"ami":13.34,"earnings22":9.31,"gigaspeech":8.8,"spgispeech":2.23,"voxpopuli":2.6,"urgent2024":17.95,"monsoon":8.86,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), mistralai/Voxtral-Mini-4B-Realtime-2602 had an average word error rate of 7.74 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-mistralai-voxtral-mini-4b-realtime-2602","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"Edge0/Audio8-ASR-0.1B","value":7.84,"low":null,"high":null,"as_written":"7.84","unit":"percent","model":"Edge0/Audio8-ASR-0.1B","maker":"Edge0","access":"Open weights","product":null,"license":"apache-2.0","sizeB":0.324,"rtfx":725.77,"lsClean":2.71,"lsOther":6.6,"ami":11,"earnings22":12.53,"gigaspeech":8.48,"spgispeech":3.73,"voxpopuli":4.39,"urgent2024":14.97,"monsoon":6.18,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), Edge0/Audio8-ASR-0.1B had an average word error rate of 7.84 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-edge0-audio8-asr-0-1b","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"abr-ai/niagara-84m-batch.en","value":8.87,"low":null,"high":null,"as_written":"8.87","unit":"percent","model":"abr-ai/niagara-84m-batch.en","maker":"abr-ai","access":"Open weights","product":null,"license":"applied-brain-research-open-license","sizeB":0.084,"rtfx":1940.83,"lsClean":3.09,"lsOther":7.6,"ami":12.01,"earnings22":10.83,"gigaspeech":10.41,"spgispeech":2.61,"voxpopuli":4.33,"urgent2024":20.16,"monsoon":8.82,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), abr-ai/niagara-84m-batch.en had an average word error rate of 8.87 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-abr-ai-niagara-84m-batch-en","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"nvidia/nemotron-3.5-asr-streaming-0.6b","value":9.24,"low":null,"high":null,"as_written":"9.24","unit":"percent","model":"nvidia/nemotron-3.5-asr-streaming-0.6b","maker":"nvidia","access":"Open weights","product":"NVIDIA Nemotron Speech ASR","license":"OpenMDW-1.1","sizeB":0.64,"rtfx":1348.36,"lsClean":2.83,"lsOther":6.79,"ami":13.43,"earnings22":15.3,"gigaspeech":9.86,"spgispeech":3.27,"voxpopuli":4.24,"urgent2024":20.14,"monsoon":7.3,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), nvidia/nemotron-3.5-asr-streaming-0.6b had an average word error rate of 9.24 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-nvidia-nemotron-3-5-asr-streaming-0-6b","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"facebook/omniASR-CTC-7B-v2","value":10.13,"low":null,"high":null,"as_written":"10.13","unit":"percent","model":"facebook/omniASR-CTC-7B-v2","maker":"facebook","access":"Open weights","product":"Omnilingual ASR","license":"apache-2.0","sizeB":6.5,"rtfx":522.26,"lsClean":2.12,"lsOther":5.05,"ami":17.44,"earnings22":18.3,"gigaspeech":12.95,"spgispeech":5.16,"voxpopuli":4.43,"urgent2024":18.68,"monsoon":7.08,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), facebook/omniASR-CTC-7B-v2 had an average word error rate of 10.13 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-facebook-omniasr-ctc-7b-v2","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"abr-ai/niagara-38m-batch.en","value":10.31,"low":null,"high":null,"as_written":"10.31","unit":"percent","model":"abr-ai/niagara-38m-batch.en","maker":"abr-ai","access":"Open weights","product":null,"license":"applied-brain-research-open-license","sizeB":0.038,"rtfx":4561.35,"lsClean":3.81,"lsOther":9.28,"ami":13.87,"earnings22":12.94,"gigaspeech":11.38,"spgispeech":3.08,"voxpopuli":5.16,"urgent2024":22.88,"monsoon":10.41,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), abr-ai/niagara-38m-batch.en had an average word error rate of 10.31 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-abr-ai-niagara-38m-batch-en","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"google/gemma-4-E4B-it","value":11.38,"low":null,"high":null,"as_written":"11.38","unit":"percent","model":"google/gemma-4-E4B-it","maker":"google","access":"Open weights","product":null,"license":"apache-2.0","sizeB":7.94,"rtfx":160.69,"lsClean":3.85,"lsOther":9.97,"ami":20.57,"earnings22":14.16,"gigaspeech":11.74,"spgispeech":5.49,"voxpopuli":4.09,"urgent2024":26.38,"monsoon":6.19,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), google/gemma-4-E4B-it had an average word error rate of 11.38 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-google-gemma-4-e4b-it","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"abr-ai/niagara-19m-batch.en","value":12.08,"low":null,"high":null,"as_written":"12.08","unit":"percent","model":"abr-ai/niagara-19m-batch.en","maker":"abr-ai","access":"Open weights","product":null,"license":"applied-brain-research-open-license","sizeB":0.02,"rtfx":4348.44,"lsClean":4.74,"lsOther":11.02,"ami":16.17,"earnings22":16.1,"gigaspeech":12.72,"spgispeech":3.67,"voxpopuli":5.73,"urgent2024":25.4,"monsoon":13.15,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), abr-ai/niagara-19m-batch.en had an average word error rate of 12.08 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-abr-ai-niagara-19m-batch-en","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"abr-ai/niagara-9m-batch.en","value":15.27,"low":null,"high":null,"as_written":"15.27","unit":"percent","model":"abr-ai/niagara-9m-batch.en","maker":"abr-ai","access":"Open weights","product":null,"license":"applied-brain-research-open-license","sizeB":0.009,"rtfx":7217.02,"lsClean":6.45,"lsOther":14.53,"ami":20.94,"earnings22":21.46,"gigaspeech":14.97,"spgispeech":5.16,"voxpopuli":7.33,"urgent2024":29.47,"monsoon":17.11,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), abr-ai/niagara-9m-batch.en had an average word error rate of 15.27 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-abr-ai-niagara-9m-batch-en","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"OtterAI/speech-v3.1","value":3.88,"low":null,"high":null,"as_written":"3.88","unit":"percent","model":"OtterAI/speech-v3.1","maker":"OtterAI","access":"Cloud API","product":"Otter.ai","license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":0.9,"lsOther":1.89,"ami":6.42,"earnings22":4.55,"gigaspeech":7.13,"spgispeech":1.56,"voxpopuli":1.42,"urgent2024":8.07,"monsoon":3.02,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), OtterAI/speech-v3.1 had an average word error rate of 3.88 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-otterai-speech-v3-1","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"zoom/scribe_v2_pro","value":4.08,"low":null,"high":null,"as_written":"4.08","unit":"percent","model":"zoom/scribe_v2_pro","maker":"zoom","access":"Cloud API","product":null,"license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":1.11,"lsOther":2.24,"ami":6.08,"earnings22":4.12,"gigaspeech":7.75,"spgispeech":1.36,"voxpopuli":3.08,"urgent2024":7.99,"monsoon":3,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), zoom/scribe_v2_pro had an average word error rate of 4.08 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-zoom-scribe-v2-pro","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"microsoft/azure-speech-07-2026","value":4.49,"low":null,"high":null,"as_written":"4.49","unit":"percent","model":"microsoft/azure-speech-07-2026","maker":"microsoft","access":"Cloud API","product":"Azure AI Speech","license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":1.23,"lsOther":2.71,"ami":7.05,"earnings22":5.2,"gigaspeech":7.48,"spgispeech":2.17,"voxpopuli":1.65,"urgent2024":9.9,"monsoon":3,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), microsoft/azure-speech-07-2026 had an average word error rate of 4.49 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-microsoft-azure-speech-07-2026","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"modulate/multilingual","value":4.53,"low":null,"high":null,"as_written":"4.53","unit":"percent","model":"modulate/multilingual","maker":"modulate","access":"Cloud API","product":"Modulate Velma Transcribe","license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":0.87,"lsOther":1.77,"ami":6.31,"earnings22":6,"gigaspeech":6.91,"spgispeech":2.06,"voxpopuli":3.55,"urgent2024":10.04,"monsoon":3.26,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), modulate/multilingual had an average word error rate of 4.53 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-modulate-multilingual","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"elevenlabs/scribe_v2","value":4.55,"low":null,"high":null,"as_written":"4.55","unit":"percent","model":"elevenlabs/scribe_v2","maker":"elevenlabs","access":"Cloud API","product":"ElevenLabs Scribe","license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":1.06,"lsOther":2.33,"ami":8.19,"earnings22":4.8,"gigaspeech":7.63,"spgispeech":3.01,"voxpopuli":1.41,"urgent2024":9.2,"monsoon":3.32,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), elevenlabs/scribe_v2 had an average word error rate of 4.55 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-elevenlabs-scribe-v2","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"sophea/asr-k1 (preview)","value":4.61,"low":null,"high":null,"as_written":"4.61","unit":"percent","model":"sophea/asr-k1 (preview)","maker":"sophea","access":"Cloud API","product":null,"license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":0.9,"lsOther":2.05,"ami":6.28,"earnings22":5.56,"gigaspeech":7.96,"spgispeech":2.11,"voxpopuli":3.35,"urgent2024":10.07,"monsoon":3.25,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), sophea/asr-k1 (preview) had an average word error rate of 4.61 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-sophea-asr-k1-preview","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"reson8/resonant-1","value":4.64,"low":null,"high":null,"as_written":"4.64","unit":"percent","model":"reson8/resonant-1","maker":"reson8","access":"Cloud API","product":"Reson8","license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":0.92,"lsOther":2.09,"ami":8.22,"earnings22":5.48,"gigaspeech":7.38,"spgispeech":2.77,"voxpopuli":2.19,"urgent2024":9.64,"monsoon":3.04,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), reson8/resonant-1 had an average word error rate of 4.64 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-reson8-resonant-1","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"assemblyai/universal-3-5-pro","value":4.87,"low":null,"high":null,"as_written":"4.87","unit":"percent","model":"assemblyai/universal-3-5-pro","maker":"assemblyai","access":"Cloud API","product":"AssemblyAI","license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":1.1,"lsOther":2.11,"ami":10.63,"earnings22":6.05,"gigaspeech":7.62,"spgispeech":1.81,"voxpopuli":1.92,"urgent2024":9.08,"monsoon":3.49,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), assemblyai/universal-3-5-pro had an average word error rate of 4.87 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-assemblyai-universal-3-5-pro","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"smallestai/pulse","value":5.08,"low":null,"high":null,"as_written":"5.08","unit":"percent","model":"smallestai/pulse","maker":"smallestai","access":"Cloud API","product":"Smallest.ai Pulse","license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":1.3,"lsOther":2.7,"ami":9.11,"earnings22":4.74,"gigaspeech":8.18,"spgispeech":1.91,"voxpopuli":3.85,"urgent2024":10.44,"monsoon":3.52,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), smallestai/pulse had an average word error rate of 5.08 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-smallestai-pulse","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"sprag/symphony","value":5.27,"low":null,"high":null,"as_written":"5.27","unit":"percent","model":"sprag/symphony","maker":"sprag","access":"Cloud API","product":null,"license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":1,"lsOther":2.19,"ami":9.02,"earnings22":6.01,"gigaspeech":7.52,"spgispeech":2.21,"voxpopuli":3.33,"urgent2024":12.64,"monsoon":3.55,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), sprag/symphony had an average word error rate of 5.27 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-sprag-symphony","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"openai/gpt-transcribe","value":5.49,"low":null,"high":null,"as_written":"5.49","unit":"percent","model":"openai/gpt-transcribe","maker":"openai","access":"Cloud API","product":"OpenAI Speech to Text API","license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":1.31,"lsOther":3.39,"ami":8.95,"earnings22":6.61,"gigaspeech":7.76,"spgispeech":2.23,"voxpopuli":2.65,"urgent2024":12.86,"monsoon":3.63,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), openai/gpt-transcribe had an average word error rate of 5.49 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-openai-gpt-transcribe","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"gladia/solaria-3","value":5.59,"low":null,"high":null,"as_written":"5.59","unit":"percent","model":"gladia/solaria-3","maker":"gladia","access":"Cloud API","product":"Gladia","license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":1.31,"lsOther":2.99,"ami":9.89,"earnings22":5.94,"gigaspeech":7.32,"spgispeech":2.68,"voxpopuli":2.76,"urgent2024":13.73,"monsoon":3.7,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), gladia/solaria-3 had an average word error rate of 5.59 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-gladia-solaria-3","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"},{"row":"meta/muse-voice-transcribe","value":5.75,"low":null,"high":null,"as_written":"5.75","unit":"percent","model":"meta/muse-voice-transcribe","maker":"meta","access":"Cloud API","product":"Meta Muse Voice Transcribe","license":"Proprietary","sizeB":null,"rtfx":null,"lsClean":1.84,"lsOther":4.51,"ami":9.2,"earnings22":6.01,"gigaspeech":7.75,"spgispeech":3.21,"voxpopuli":2.02,"urgent2024":13.03,"monsoon":4.16,"version":"2026-10-17","conversion":null,"claim":"On the Hugging Face Open ASR Leaderboard (its version of October 17, 2026, read on October 20, 2026), meta/muse-voice-transcribe had an average word error rate of 5.75 percent across the leaderboard's 9 public English test sets.","quote":null,"where_in_source":null,"fact_url":"https://yapapp.cc/facts/t-open-asr-leaderboard-meta-muse-voice-transcribe","source_title":null,"source_publisher":null,"source_year":null,"source_url":null,"source_doi":null,"checked":"2026-10-20"}]}