| 1 | wav2vec2-ljspeech-gruut | — | bookbot | 100.0 | low | |
| 2 | mms-300m-waxal-lug | facebook/mms-300m | waxal-benchmarking | 100.0 | low | |
| 3 | mms-300m-waxal-sna | facebook/mms-300m | waxal-benchmarking | 100.0 | low | |
| 4 | mms-300m-waxal-lin | facebook/mms-300m | waxal-benchmarking | 100.0 | low | |
| 5 | w2vbert-luganda-waxal-punct | sulaimank/w2vbert-luganda-waxal-stage2 | sulaimank | 100.0 | low | |
| 6 | mms-300m-waxal-ach | facebook/mms-300m | waxal-benchmarking | 100.0 | low | |
| 7 | phoneme-scorer-v2-wav2vec2 | — | ct-vikramanantha | 100.0 | low | |
| 8 | mms-300m-waxal-nyn | facebook/mms-300m | waxal-benchmarking | 100.0 | low | |
| 9 | mms-300m-waxal-mas | facebook/mms-300m | waxal-benchmarking | 100.0 | low | |
| 10 | whisper-small-lingala-qlora | openai/whisper-small | Elod1e | 99.73 | medium | |
| 11 | Phi-4-multimodal-instruct | microsoft/Phi-4-multimodal-instruct | microsoft | 99.63 | low | fleurs-ko2en-cot | fleurs-en2ko-cot |
| 12 | Phi-4-multimodal-instruct | microsoft/Phi-4-multimodal-instruct | Lexius | 99.63 | low | fleurs-ko2en-cot | fleurs-en2ko-cot |
| 13 | whisper-small-finetuned-iwslt-irish-welsh-curriculum-synthetic-b | openai/whisper-small | jjnhuang | 98.0 | low | |
| 14 | whisper-small-finetuned-iwslt-irish-welsh-50-50-synthetic | openai/whisper-small | jjnhuang | 95.92 | medium | |
| 15 | whisper-small-finetuned-iwslt-irish-transfer-welsh-synthetic | openai/whisper-small | jjnhuang | 95.26 | medium | |
| 16 | wav2vec2-large-xlsr-53-th | — | airesearch | 95.24 | low | |
| 17 | whisper-small-finetuned-iwslt-irish-synthetic | openai/whisper-small | jjnhuang | 94.94 | medium | |
| 18 | whisper-small-finetuned-iwslt-irish-welsh-curriculum-synthetic | openai/whisper-small | jjnhuang | 93.83 | medium | |
| 19 | moonshine-tiny-vi | UsefulSensors/moonshine-tiny-vi | UsefulSensors | 91.89 | low | |
| 20 | moonshine-tiny-vi-ONNX | UsefulSensors/moonshine-tiny-vi | onnx-community | 91.89 | low | |
| 21 | whisper-tiny-ja | openai/whisper-tiny | zubokol | 90.98 | medium | |
| 22 | cali-whisper-tiny.en-drop-004-production | openai/whisper-tiny | 1Developer | 87.12 | low | |
| 23 | cali-whisper-tiny.en-drop-003-production | openai/whisper-tiny | 1Developer | 87.12 | low | |
| 24 | whisper-medium-hindi-hinglish-asr-fine-tuned | openai/whisper-medium | shujaAK | 86.99 | medium | Relative WER Improvement |
| 25 | whisper-medium-hindi-hinglish-asr-fine-tuned | openai/whisper-medium | shujaAK | 86.99 | medium | WER Improvement |
| 26 | ATOM | — | Chillarmo | 86.6 | low | |
| 27 | wav2vec2-large-xlsr-tamil-commonvoice | — | nikhil6041 | 85.12 | medium | |
| 28 | whisper-small-yemeni | openai/whisper-small | tahaalselwii | 84.32 | low | |
| 29 | wav2vec2-xlsr-multilingual-56 | — | voidful | 82.5 | high | |
| 30 | wav2vec2-large-xlsr-53-chinese-zh-cn | jonatasgrosman/wav2vec2-large-xlsr-53-chinese-zh-cn | jonatasgrosman | 82.37 | high | |
| 31 | parakeet-tdt-0.6b-v3 | nvidia/parakeet-tdt-0.6b-v3 | sonic-speech | 82.0 | low | |
| 32 | parakeet-tdt-0.6b-v3-int8 | sonic-speech/parakeet-tdt-0.6b-v3 | sonic-speech | 82.0 | medium | WER (LibriSpeech) |
| 33 | wav2vec2-large-xlsr-53-japanese | jonatasgrosman/wav2vec2-large-xlsr-53-japanese | jonatasgrosman | 81.8 | high | |
| 34 | nemotron-3.5-nb-NO | nvidia/nemotron-3.5-asr-streaming-0.6b | LokaalHub | 76.09 | medium | WER (offline, full-context, normalized) on `LokaalHub/nb-NO-asr-cv` test |
| 35 | qwen3-asr-0.6b-arabic-dialectal | Qwen/Qwen3-ASR-0.6B-hf | oddadmix | 75.6 | high | Zero-shot WER |
| 36 | wav2vec2-base-german | facebook/wav2vec2-base | aware-ai | 74.28 | medium | |
| 37 | wav2vec2-large-japanese | — | NTQAI | 73.1 | high | |
| 38 | whisper-algerian-darja-small | openai/whisper-small | touati-kamel | 71.73 | low | |
| 39 | wav2vec2-large-chinese-zh-cn | — | wbbbbb | 70.47 | high | |
| 40 | wav2vec2-large-xls-r-300m-armenian | — | infinitejoy | 69.42 | medium | |
| 41 | whisper-large-v3-turbo-singlish | openai/whisper-large-v3-turbo | mjwong | 68.79 | low | |
| 42 | wav2vec2-large-mms-1b-yoruba-test | facebook/mms-1b-all | ogbi | 68.02 | medium | |
| 43 | whisper-tiny-yoruba1 | EYEDOL/whisper-tiny-yoruba | EYEDOL | 67.55 | medium | |
| 44 | whisper-small-burmese-v2 | myatsu/whisper-small-burmese | myatsu | 67.36 | medium | |
| 45 | whisper-tiny-grpo-5e2f49e-cv22-tiny-sftadaptive | openai/whisper-tiny | wrice | 67.1 | medium | |
| 46 | moulsot.v0.3 | Qwen/Qwen3-ASR-1.7B | atlasia | 66.78 | medium | |
| 47 | moonshine-base-bn | UsefulSensors/moonshine-base | Sanjidh090 | 66.71 | low | |
| 48 | whisper-tiny-yoruba2 | EYEDOL/whisper-tiny-yoruba1 | EYEDOL | 66.05 | medium | |
| 49 | moonshine-tiny-ar-ONNX | UsefulSensors/moonshine-tiny-ar | onnx-community | 66.01 | low | |
| 50 | moonshine-tiny-ar | UsefulSensors/moonshine-tiny-ar | UsefulSensors | 66.01 | low | |
| 51 | whisper-large-v3_finetuned_ugandan_luganda_nonstandard_speech_v1.0 | — | cdli | 66.0 | low | |
| 52 | cdli-qwen3-asr-lg-atypical-stage3-1p7b-base | KasuleTrevor/cdli-qwen3-asr-lg-typical-1p7b-base-finetune | KasuleTrevor | 65.91 | medium | |
| 53 | whisper-tiny-grpo-4bddbd7-cv22-tiny-klbeta02 | openai/whisper-tiny | wrice | 65.6 | medium | |
| 54 | MSP-Visual | — | MahmoodAnaam | 64.93 | medium | |
| 55 | wav2vec2-large-xls-r-300m-assamese | facebook/wav2vec2-xls-r-300m | jayashreedevi2020 | 64.7 | medium | |
| 56 | whisper-small-khmer | — | seanghay | 64.64 | medium | |
| 57 | moonshine-tiny-uk-ONNX | UsefulSensors/moonshine-tiny-uk | onnx-community | 63.83 | low | |
| 58 | moonshine-tiny-uk | UsefulSensors/moonshine-tiny-uk | UsefulSensors | 63.83 | low | |
| 59 | whisper-small-khmer-v2 | — | seanghay | 61.65 | medium | |
| 60 | whisper-algerian-darja-medium | openai/whisper-medium | touati-kamel | 61.42 | low | |
| 61 | moonshine-algerian-darja-base | UsefulSensors/moonshine-base | touati-kamel | 61.42 | low | |
| 62 | swahili | facebook/wav2vec2-base | etonkou | 60.55 | medium | |
| 63 | wav2vec2-large-xls-r-300m-marathi-cv8 | — | infinitejoy | 60.49 | medium | |
| 64 | whisper-large-v3-turbo-arabic-dialectal | openai/whisper-large-v3-turbo | oddadmix | 59.0 | low | Zero-shot WER |
| 65 | cohere-transcribe-arabic-07-2026-dialectal | CohereLabs/cohere-transcribe-arabic-07-2026 | oddadmix | 59.0 | low | Zero-shot WER |
| 66 | english-filipino-wav2vec2-l-xls-r-test-09 | — | Khalsuu | 57.5 | medium | |
| 67 | afrivoices-asr-baobab | — | Okpeyemi | 56.6 | low | |
| 68 | monsoon-whisper-medium-gigaspeech2 | — | typhoon-ai | 55.64 | high | |
| 69 | qwen3-asr-0.6b-arabic-dialectal-v2 | Qwen/Qwen3-ASR-0.6B-hf | oddadmix | 55.14 | low | |
| 70 | whisper-large-v3-myanmar | openai/whisper-large-v3 | chuuhtetnaing | 54.9 | medium | |
| 71 | whisper_small_ps_augmented | ihanif/whisper_small_ps_augmented | ihanif | 53.62 | medium | |
| 72 | asr-cherokee | facebook/wav2vec2-large-xlsr-53 | charliemcvicker | 53.01 | medium | |
| 73 | whisper-small-p0km | openai/whisper-small | rick-c831 | 52.73 | medium | |
| 74 | azerbaijani-whisper-turbo | openai/whisper-large-v3-turbo | LocalDoc | 52.17 | low | |
| 75 | azerbaijani-whisper-small | openai/whisper-small | LocalDoc | 52.17 | high | |
| 76 | wav2vec2-xls-r-300m-pashto | facebook/wav2vec2-xls-r-300m | ihanif | 51.59 | medium | |
| 77 | whisper-tiny-hausa1 | EYEDOL/whisper-tiny-hausa | EYEDOL | 50.24 | medium | |
| 78 | cdli-whisper-ml-lug-swa-sunbird-full-a40 | Sunbird/asr-whisper-large-v3-salt | KasuleTrevor | 50.22 | medium | |
| 79 | whisper-small-shona | openai/whisper-small | steja | 49.91 | medium | |
| 80 | whisper-large-v3-turbo-lt | openai/whisper-large-v3-turbo | svogunas | 49.66 | low | |
| 81 | wav2vec2-xls-r-300m-sk-cv8 | — | comodoro | 49.57 | medium | |
| 82 | mms-300m-waxal-mas | facebook/mms-300m | waxal-benchmarking | 49.5 | medium | **Test WER** |
| 83 | mms-300m-waxal-mas | facebook/mms-300m | waxal-benchmarking | 49.5 | medium | **WER** |
| 84 | wav2vec2-large-xlsr-53-punjabi | Harveenchadha/vakyansh-wav2vec2-punjabi-pam-10 | kingabzpro | 49.39 | medium | |
| 85 | whisper-small-myanmar | openai/whisper-small | chuuhtetnaing | 49.06 | medium | |
| 86 | whisper-tiny-waxal-lin | openai/whisper-tiny | waxal-benchmarking | 49.0 | medium | **Test WER** |
| 87 | whisper-tiny-waxal-lin | openai/whisper-tiny | waxal-benchmarking | 49.0 | medium | **WER** |
| 88 | whisper-small-darija | openai/whisper-small | ychafiqui | 48.27 | medium | |
| 89 | whisper-tiny-hausa2 | EYEDOL/whisper-tiny-hausa1 | EYEDOL | 48.09 | medium | |
| 90 | whisper-tiny-mla-24lang | openai/whisper-tiny | burakaydinofficial | 47.6 | high | |
| 91 | Sunflower-Gemma4-E2B | google/gemma-4-E2B-it | Sunbird | 47.2 | low | |
| 92 | whisper-small-hu | openai/whisper-small | SubZtep | 46.79 | low | |
| 93 | wav2vec2-large-xls-r-300m-bulgarian | — | infinitejoy | 46.74 | medium | |
| 94 | whisper-small-sinhala_v3 | openai/whisper-small | Lingalingeswaran | 46.46 | medium | |
| 95 | whisper-tiny-waxal-lin | openai/whisper-tiny | waxal-benchmarking | 46.13 | low | |
| 96 | whisper-base-bn | openai/whisper-base | shhossain | 46.0 | medium | |
| 97 | cohere-transcribe-arabic-07-2026-int8 | CohereLabs/cohere-transcribe-arabic-07-2026 | NAMAA-Space | 46.0 | low | |
| 98 | cdli-whisper-ml-lug-swa-sunbird-full-a40-lr1e4 | Sunbird/asr-whisper-large-v3-salt | KasuleTrevor | 45.48 | medium | |
| 99 | whisper-tiny-oilgas-russian | openai/whisper-tiny | N07P | 45.08 | medium | |
| 100 | whisper-small-waxal-ful | openai/whisper-small | waxal-benchmarking | 44.82 | low | |
| 101 | whisper-small.en-atco2-asr | openai/whisper-small.en | jlvdoorn | 44.66 | medium | |
| 102 | whisper-tiny.en | openai/whisper-tiny.en | eai6 | 44.35 | medium | |
| 103 | whisper-wolof | openai/whisper-small | cibfaye | 43.94 | medium | |
| 104 | STT-Whisper-Pashto | openai/whisper-small | Sabtain-Dev | 43.83 | medium | |
| 105 | cdli-whisper-ml-lug-swa-sunbird-full-a40-lr5e5-5k | Sunbird/asr-whisper-large-v3-salt | KasuleTrevor | 43.44 | medium | |
| 106 | wav2vec2-large-mms-1b-somalia | facebook/mms-1b-all | skydheere | 43.3 | medium | |
| 107 | whisper-small-arabic-dialectal | openai/whisper-small | oddadmix | 42.8 | high | Zero-shot WER |
| 108 | whisper-small-waxal-lin | openai/whisper-small | waxal-benchmarking | 42.7 | medium | **Test WER** |
| 109 | whisper-small-waxal-lin | openai/whisper-small | waxal-benchmarking | 42.7 | medium | **WER** |
| 110 | mms-300m-waxal-lin | facebook/mms-300m | waxal-benchmarking | 42.6 | medium | **Test WER** |
| 111 | mms-300m-waxal-lin | facebook/mms-300m | waxal-benchmarking | 42.6 | medium | **WER** |
| 112 | whisper-small-waxal-ful | openai/whisper-small | waxal-benchmarking | 42.6 | medium | **Test WER** |
| 113 | whisper-small-waxal-ful | openai/whisper-small | waxal-benchmarking | 42.6 | medium | **WER** |
| 114 | Ethio-ASR-tigrinya | facebook/w2v-bert-2.0 | badrex | 42.42 | low | |
| 115 | mms-300m-waxal-ach | facebook/mms-300m | waxal-benchmarking | 42.3 | medium | **Test WER** |
| 116 | mms-300m-waxal-ach | facebook/mms-300m | waxal-benchmarking | 42.3 | medium | **WER** |
| 117 | whisper-small-zh_tw | — | Jingmiao | 41.97 | medium | |
| 118 | griot-nano-1 | — | Qlerqly | 41.67 | low | |
| 119 | griot-nano-1-onnx | — | Qlerqly | 41.67 | low | |
| 120 | wav2vec2-large-xlsr-53-finnish | — | jonatasgrosman | 41.6 | high | |
| 121 | BlueMagpie-TTS | OpenFormosa/BlueMagpie-TTS | OpenFormosa | 41.48 | low | |
| 122 | whisper-small-he | openai/whisper-small | tomtom5 | 41.34 | medium | |
| 123 | whisper-large-v3-turbo-swiss-german | openai/whisper-large-v3-turbo | Flurin17 | 41.05 | low | |
| 124 | whisper-small-arabic-dialectal-v2 | openai/whisper-small | oddadmix | 40.34 | low | |
| 125 | VoiceTut-TTS | k2-fsa/OmniVoice | mohammedaly22 | 40.0 | medium | WER — Egyptian Arabic |
| 126 | whisper-small-waxal-lin | openai/whisper-small | waxal-benchmarking | 39.94 | low | |
| 127 | wav2vec2-base-it-latin | facebook/wav2vec2-base | lsb | 39.8 | medium | |
| 128 | cdli-whisper-ml-ugeng-lug-swa-sunbird-full-a40-lr1e4-5k | Sunbird/asr-whisper-large-v3-salt | KasuleTrevor | 39.61 | medium | |
| 129 | wav2vec2-large-xlsr-53-arabic | jonatasgrosman/wav2vec2-large-xlsr-53-arabic | jonatasgrosman | 39.59 | high | |
| 130 | whisper-medium-ml | openai/whisper-medium | thennal | 38.62 | medium | |
| 131 | mms-300m-waxal-nyn | facebook/mms-300m | waxal-benchmarking | 38.6 | medium | **Test WER** |
| 132 | mms-300m-waxal-nyn | facebook/mms-300m | waxal-benchmarking | 38.6 | medium | **WER** |
| 133 | xls-r-uzbek-cv8 | facebook/wav2vec2-xls-r-300m | lucio | 38.52 | medium | |
| 134 | cdli-whisper-ml-ugeng-lug-swa-sunbird-full-a40-lr5e5 | Sunbird/asr-whisper-large-v3-salt | KasuleTrevor | 37.97 | medium | |
| 135 | wav2vec2-child-en-tokenizer-4 | — | jaeyeon | 37.69 | medium | |
| 136 | whisper-small-malayalam | openai/whisper-small | sajilck | 37.64 | medium | |
| 137 | my_zh_CN_asr_cv13_model | jonatasgrosman/wav2vec2-large-xlsr-53-chinese-zh-cn | tristayqc | 37.5 | medium | |
| 138 | wav2vec2-base-timit-ms | facebook/wav2vec2-base | saribalgar | 37.22 | medium | |
| 139 | whisper-base-mla-24lang | openai/whisper-base | burakaydinofficial | 37.0 | high | |
| 140 | whisper-small-waxal-sna | openai/whisper-small | waxal-benchmarking | 36.3 | low | |
| 141 | mms-1b-all-indic_voices-normalized-30_70-lg-hin-steps-10000-lr-0.0001-FT | facebook/mms-1b-all | dianavdavidson | 36.09 | medium | |
| 142 | w2vbert-waxal-multi7 | sulaimank/w2v-bert-waxal | sulaimank | 35.88 | medium | |
| 143 | Wolof-HuBERT-CTC | soynade-research/Wolof-HuBERT-Base | soynade-research | 35.65 | medium | |
| 144 | w2v-bert-2.0-afrivoices-ea-6l-asr-int8-edge | leophill/w2v-bert-2.0-afrivoices-ea-6l-asr | leophill | 34.81 | medium | |
| 145 | wav2vec2-large-xlsr-lithuanian | — | m3hrdadfi | 34.66 | medium | |
| 146 | whisper-small-uz-v1 | openai/whisper-small | OvozifyLabs | 34.5 | high | |
| 147 | whisper-small-combined-lin-sna-lug-olusola | openai/whisper-small | samolusola-personal | 34.43 | medium | |
| 148 | whisper-large-v3-turbo-arabic-dialectal | openai/whisper-large-v3-turbo | oddadmix | 34.4 | medium | |
| 149 | cohere-transcribe-arabic-07-2026-dialectal | CohereLabs/cohere-transcribe-arabic-07-2026 | oddadmix | 34.4 | medium | |
| 150 | whisper-small-arabic-dialectal | openai/whisper-small | oddadmix | 34.4 | medium | |
| 151 | qwen3-asr-0.6b-arabic-dialectal | Qwen/Qwen3-ASR-0.6B-hf | oddadmix | 34.4 | medium | |
| 152 | whisper-medium-combined-lin-sna-lug-olusola | openai/whisper-medium | samolusola-personal | 34.36 | medium | |
| 153 | whisper-tiny-waxal-lug | openai/whisper-tiny | waxal-benchmarking | 33.8 | medium | **Test WER** |
| 154 | whisper-tiny-waxal-lug | openai/whisper-tiny | waxal-benchmarking | 33.8 | medium | **WER** |
| 155 | wav2vec2-xls-r-300m-hy | — | arampacha | 33.33 | medium | |
| 156 | whisper-large-v3-turbo-arabic-dialectal-v2 | openai/whisper-large-v3-turbo | oddadmix | 33.24 | low | |
| 157 | whisper-base-armenian | openai/whisper-base | Chillarmo | 33.19 | medium | **Word Error Rate (WER)** |
| 158 | whisper-base-ru | ovsyannikovls/whisper-base-ru | ovsyannikovls | 33.1 | medium | |
| 159 | swahili-wav2vec2-asr | eddiegulay/wav2vec2-large-xlsr-mvc-swahili | RareElf | 33.0 | medium | WER |
| 160 | mms-1b-zdj | facebook/mms-1b-all | rifailabs | 32.96 | medium | |
| 161 | whisper-tiny-waxal-lug | openai/whisper-tiny | waxal-benchmarking | 32.87 | low | |
| 162 | whisper-small-pashto | ihanif/whisper_small_ps_augmented | Adnan666 | 32.6 | medium | |
| 163 | uz-whisper-small-stt-v2 | maqsudxo1ja/uz-whisper-small-stt | maqsudxo1ja | 32.49 | low | |
| 164 | whisper-tiny-en-minds14 | openai/whisper-tiny | nagorebravo | 32.41 | medium | |
| 165 | pashto-whisper-small | openai/whisper-small | karimnaimy | 32.34 | medium | |
| 166 | wav2vec2-large-xlsr-persian | — | m3hrdadfi | 32.2 | medium | |
| 167 | whisper-small-hi | openai/whisper-small | sanchit-gandhi | 32.01 | medium | |
| 168 | whisper-large-v3-arabic-dialectal-v2 | openai/whisper-large-v3 | oddadmix | 31.99 | low | |
| 169 | whisper-small-persian-v1 | openai/whisper-small | aliyzd95 | 31.93 | medium | |
| 170 | wav2vec2-large-xlsr-53-hungarian | — | jonatasgrosman | 31.4 | high | |
| 171 | whisper-large-libyan | openai/whisper-large-v3 | Rhaodgh | 31.36 | medium | |
| 172 | whisper-large-v3-turbo-arabic | openai/whisper-large-v3-turbo | mboushaba | 31.15 | medium | |
| 173 | Ethio-ASR-amharic | facebook/w2v-bert-2.0 | badrex | 30.95 | low | |
| 174 | whisper-large-v3-bn-cv17 | openai/whisper-large-v3 | bijaykumarsingh | 30.84 | medium | |
| 175 | wav2vec2-large-xlsr-japanese | — | vumichien | 30.84 | medium | |
| 176 | ASR-Somali | — | Mustafaa4a | 30.6 | medium | |
| 177 | code-switched-egyptian-arabic-whisper-small | openai/whisper-small | IbrahimAmin | 30.29 | low | |
| 178 | wav2vec2-large-xlsr-53-persian | — | jonatasgrosman | 30.12 | high | |
| 179 | wav2vec2-large-xls-r-300m-bg | — | anuragshas | 30.02 | medium | |
| 180 | uk_wav2vec2_with_stress_mark | mouseyy/result_data_2-3 | mouseyy | 29.84 | medium | |
| 181 | jem-whisper-yi-2026-07-24 | ivrit-ai/yi-whisper-large-v3 | Kohn-AI | 29.83 | medium | |
| 182 | nemotron-3.5-swahili-streaming-asr | nvidia/nemotron-3.5-asr-streaming-0.6b | Tonykip | 29.8 | low | |
| 183 | wav2vec2-xls-r-300m-korean | — | w11wo | 29.54 | low | |
| 184 | filipino-wav2vec2-l-xls-r-300m-official | — | Khalsuu | 29.22 | medium | |
| 185 | whisper-small-si | openai/whisper-small | seniruk | 29.14 | medium | |
| 186 | uz-whisper-small-stt-v2 | maqsudxo1ja/uz-whisper-small-stt | maqsudxo1ja | 28.64 | medium | **WER (Word Error Rate)** |
| 187 | wav2vec2-large-xls-r-300m-bg-d2 | — | DrishtiSharma | 28.6 | medium | |
| 188 | gha-whisper-small-twi-v3 | teckedd/whisper-small-waxal-round2-specaug-v1 | teckedd | 28.07 | medium | |
| 189 | xlsr-waxal-from-luganda | sulaimank/xls-r-300m-luganda-313hr | sulaimank | 28.06 | medium | |
| 190 | whisper-small-yoruba | openai/whisper-small | LyngualLabs | 27.92 | low | |
| 191 | pashto-voice-v2 | openai/whisper-small | karimnaimy | 27.7 | medium | |
| 192 | wav2vec2-large-xlsr-turkish | — | m3hrdadfi | 27.51 | medium | |
| 193 | whisper-large-v2-malayalam | — | DrishtiSharma | 27.46 | medium | |
| 194 | egyptian-arabic-wav2vec2-xlsr-53 | omarxadel/wav2vec2-large-xlsr-53-arabic-egyptian | IbrahimAmin | 27.2 | low | |
| 195 | moonshine-tiny-ar-gguf | UsefulSensors/moonshine-tiny-ar | handy-computer | 27.11 | low | |
| 196 | wav2vec2-xls-r-latin | facebook/wav2vec2-xls-r-300m | njand | 27.09 | low | |
| 197 | wav2vec2-large-xls-r-300m-welsh | — | infinitejoy | 27.02 | medium | |
| 198 | wav2vec2-large-mms-1b-all-lingala-ojpl | facebook/mms-1b-all | BrainTheos | 26.98 | medium | |
| 199 | whisper-medium-urdu | — | ihanif | 26.98 | medium | |
| 200 | whisper-small-waxal-sna | openai/whisper-small | waxal-benchmarking | 26.9 | medium | **Test WER** |
| 201 | whisper-small-waxal-sna | openai/whisper-small | waxal-benchmarking | 26.9 | medium | **WER** |
| 202 | whisper-meduim-mongolian | openai/whisper-medium | Cafet | 26.87 | medium | |
| 203 | whisper-large-v3-persian | openai/whisper-large-v3 | nezamisafa | 26.54 | medium | |
| 204 | whisper-large-persian | — | steja | 26.37 | medium | |
| 205 | wav2vec2-large-mms-1b-azerbaijani-common_voice15.0 | facebook/mms-1b-all | nijatzeynalov | 26.32 | medium | |
| 206 | whisper-large-v3-turbo-korean | openai/whisper-large-v3-turbo | ghost613 | 26.26 | medium | |
| 207 | whisper-large-v3-turbo-urdu | openai/whisper-large-v3-turbo | kingabzpro | 26.23 | medium | **WER** |
| 208 | whisper-large-v3-urdu | openai/whisper-large-v3 | kingabzpro | 26.02 | medium | **WER** |
| 209 | whisper-large-v3-eu | openai/whisper-large-v3 | xezpeleta | 25.89 | low | |
| 210 | whisper-small-persian | openai/whisper-small | AmirMohseni | 25.82 | medium | |
| 211 | whisper-large-v3-turbo-urdu | openai/whisper-large-v3-turbo | kingabzpro | 25.78 | medium | |
| 212 | whisper-small-yoruba-07-17 | openai/whisper-small | ccibeekeoc42 | 25.68 | medium | |
| 213 | MSP-Audio | facebook/wav2vec2-large-robust-ft-libri-960h | MahmoodAnaam | 25.66 | medium | |
| 214 | flix-swissgerman-full | — | Flix-AI | 25.6 | medium | WER (measured) |
| 215 | w2v-bert-2.0-khmer | facebook/w2v-bert-2.0 | Prakmlis | 25.59 | medium | |
| 216 | whisper-base-bungoma.en | openai/whisper-base.en | eai6 | 25.28 | medium | |
| 217 | tamil-asr-qwen3 | Qwen/Qwen3-ASR-1.7B | osmapi | 25.27 | low | |
| 218 | nemotron-3.5-asr-streaming-slovenian | nvidia/nemotron-3.5-asr-streaming-0.6b | texdata | 25.07 | low | |
| 219 | mms-300m-waxal-sna | facebook/mms-300m | waxal-benchmarking | 25.0 | medium | **Test WER** |
| 220 | mms-300m-waxal-sna | facebook/mms-300m | waxal-benchmarking | 25.0 | medium | **WER** |
| 221 | wav2vec2-xlsr-khmer | — | gagan3012 | 24.96 | medium | |
| 222 | whisper-large-v3-turbo-darija | openai/whisper-large-v3-turbo | anaszil | 24.9 | medium | |
| 223 | whisper-large-v3-turbo-darija | openai/whisper-large-v3-turbo | anaszil | 24.88 | medium | **Word Error Rate (WER)** |
| 224 | Audar-ASR-V1-Flash | — | audarai | 24.78 | high | |
| 225 | Audar-ASR-V1-Turbo | — | audarai | 24.78 | low | |
| 226 | wav2vec2-large-xlsr-japanese-hiragana | — | vumichien | 24.74 | medium | |
| 227 | moonshine_tiny_pt_v05 | aomocelin/moonshine_tiny_pt_v04 | aomocelin | 24.74 | medium | |
| 228 | whisper-large-v3-vaani-hindi | openai/whisper-small | ARTPARK-IISc | 24.66 | high | |
| 229 | whisper-large-v3-turbo-russian-codeswitch | coriollon/whisper-large-v3-turbo-russian | coriollon | 24.64 | medium | WER |
| 230 | whisper-large-v2-vietnamese-lyrics-transcription | openai/whisper-large-v2 | kelvinbksoh | 24.61 | high | |
| 231 | wav2vec2-bloom-speech-tgl | — | sil-ai | 24.57 | medium | |
| 232 | moonshine-base-ar-gguf | UsefulSensors/moonshine-base-ar | handy-computer | 24.45 | low | |
| 233 | Cohere-Speech-Tashkeel-2B | CohereLabs/cohere-transcribe-arabic-07-2026 | NAMAA-Space | 24.35 | medium | WER — fully diacritized |
| 234 | whisper-small-mla-24lang | openai/whisper-small | burakaydinofficial | 24.2 | high | |
| 235 | openai_whisper-large-v3-turbo-NVFP4 | openai/whisper-large-v3-turbo | Reza2kn | 24.18 | low | |
| 236 | Medical-Whisper-Large-v3 | openai/whisper-large-v3 | Na0s | 24.0 | low | |
| 237 | whisper-large-v3_finetuned_ugandan_english_nonstandard_speech_v1.0 | — | cdli | 24.0 | low | |
| 238 | whisper-large-v3-swahili-afrivoice-cv | openai/whisper-large-v3 | ElizabethMwangi | 23.93 | medium | |
| 239 | whisper-large-v3-afrikaans | openai/whisper-large-v3 | andreoosthuizen | 23.92 | low | |
| 240 | whisper-large-v3-turbo-nepali | openai/whisper-large-v3-turbo | kiranpantha | 23.63 | medium | |
| 241 | MSP-Fusion | — | MahmoodAnaam | 23.5 | medium | |
| 242 | whosper-large-v2 | openai/whisper-large-v2 | CAYTU | 23.45 | medium | WER |
| 243 | whisper_small-fa_v03 | openai/whisper-small | mohammadh128 | 23.15 | medium | |
| 244 | parakeet-tdt-0.6b-EN-Medical | nvidia/parakeet-tdt-0.6b-v3 | yuriyvnv | 22.84 | medium | |
| 245 | whisper-small-egyptian-arabic | — | MAdel121 | 22.79 | medium | |
| 246 | whisper-small-hy-2 | openai/whisper-small | ErikMkrtchyan | 22.79 | medium | |
| 247 | whisper-small-egyptian-arabic | — | MAdel121 | 22.69 | medium | WER |
| 248 | wav2vec2-base-760-turkish | facebook/wav2vec2-base | ceyda | 22.6 | medium | |
| 249 | asr-wav2vec2-commonvoice-14-pt | — | speechbrain | 22.41 | low | |
| 250 | whisper-medium-full-clinical-r16 | openai/whisper-medium | robello2 | 22.4 | medium | |
| 251 | whisper-base-google-fleurs-pt-br | openai/whisper-base | thiagobarbosa | 22.3 | medium | |
| 252 | whisper-large-v3-atco2-asr-atcosim | openai/whisper-large-v3 | jlvdoorn | 22.27 | medium | |
| 253 | whisper-large-v3-turbo-hindi-lora | openai/whisper-large-v3-turbo | Tachyeon | 22.25 | medium | |
| 254 | wav2vec2-large-xls-r-300m-Urdu | facebook/wav2vec2-xls-r-300m | kingabzpro | 22.22 | high | |
| 255 | whisper-small-waxal-lug | openai/whisper-small | waxal-benchmarking | 22.15 | low | |
| 256 | whisper-small-mn | openai/whisper-small | Blgn94 | 22.1 | medium | WER |
| 257 | whisper-large-fa-v1 | openai/whisper-large-v3-turbo | vhdm | 22.07 | low | |
| 258 | moonshine-tiny-fr-ONNX | Cornebidouil/moonshine-tiny-fr | onnx-community | 21.8 | low | |
| 259 | moonshine-tiny-fr-ONNX | Cornebidouil/moonshine-tiny-fr | onnx-community | 21.8 | medium | **Word Error Rate (WER)** |
| 260 | wav2vec2-xls-r-300m-bengali | — | arijitx | 21.72 | medium | |
| 261 | MSP | — | MahmoodAnaam | 21.6 | medium | |
| 262 | whisper-small-waxal-lug | openai/whisper-small | waxal-benchmarking | 21.6 | medium | **Test WER** |
| 263 | whisper-small-waxal-lug | openai/whisper-small | waxal-benchmarking | 21.6 | medium | **WER** |
| 264 | whisper-large-v3-urdu | openai/whisper-large-v3 | kingabzpro | 21.47 | medium | |
| 265 | xlsr-luganda-waxal | sulaimank/wav2vec2-xlsr-swahili-400hr | sulaimank | 21.44 | medium | |
| 266 | wav2vec2-xls-r-300m-wolof-lm | — | abdouaziiz | 21.25 | medium | |
| 267 | whisper-tiny_ro-80mel | openai/whisper-tiny | IonGrozea | 21.2 | medium | WER |
| 268 | w2v-bert-2.0-khmer-v4 | facebook/w2v-bert-2.0 | Prakmlis | 21.16 | medium | |
| 269 | whisper-malagasy-medium | openai/whisper-medium | Flo976 | 20.78 | medium | |
| 270 | whisper-small-yoruba | openai/whisper-small | LyngualLabs | 20.76 | medium | **Test WER** |
| 271 | whisper-tiny_ro-80mel | openai/whisper-tiny | IonGrozea | 20.73 | medium | WER (normalized: locase, punctuation stripped) |
| 272 | wav2vec2-xls-r-300m-ftspeech | facebook/wav2vec2-xls-r-300m | saattrupdan | 20.48 | low | |
| 273 | whisper-large-v3-high-mixed-nl | openai/whisper-large-v3 | yuriyvnv | 20.29 | medium | **Test WER (MLS)** |
| 274 | w2v-xls-r-uk | facebook/wav2vec2-xls-r-300m | Yehor | 20.24 | medium | |
| 275 | audioX-south-v1 | — | jiviai | 20.1 | low | |
| 276 | w2vbert-luganda-waxal-punct-v2 | sulaimank/w2vbert-luganda-waxal-stage2 | sulaimank | 20.07 | low | |
| 277 | whisper-large-v2-mn-13 | — | bayartsogt | 20.02 | medium | |
| 278 | Omnilingual-ASR-CTC-300M-CoreML-INT8-10s | facebook/omniASR-CTC-300M | aufklarer | 20.0 | low | |
| 279 | Omnilingual-ASR-CTC-300M-MLX-4bit | facebook/omniASR-CTC-300M | aufklarer | 20.0 | low | |
| 280 | whisper-medium-zulu | openai/whisper-medium | TheirStory | 19.93 | medium | |
| 281 | wav2vec2-large-mms-1b-hungarian | facebook/mms-1b-all | GaborMadarasz | 19.93 | medium | |
| 282 | Whisper-Large-v3-turbo-STT-Zeroth-KO-v2 | openai/whisper-large-v3-turbo | o0dimplz0o | 19.91 | medium | |
| 283 | QwenCleo-ASR | Qwen/Qwen3-ASR-1.7B | mohammedaly22 | 19.85 | low | |
| 284 | openai_whisper-large-v3-NVFP4 | openai/whisper-large-v3 | Reza2kn | 19.74 | low | |
| 285 | whisper-tiny-es-mix-norm | openai/whisper-tiny | deepdml | 19.6 | medium | |
| 286 | SraVaani-1.0 | — | ARTPARK-IISc | 19.4 | low | |
| 287 | Test | — | AbelWa | 19.4 | low | |
| 288 | parakeet-tdt-0.6b-EN-Medical | nvidia/parakeet-tdt-0.6b-v3 | yuriyvnv | 19.22 | low | Zero-shot WER | Zero-shot CER |
| 289 | cohere-transcribe-arabic-07-2026-dialectal-v2 | CohereLabs/cohere-transcribe-arabic-07-2026 | oddadmix | 19.1 | low | |
| 290 | whisper-small_finetuned_ghanian_ga_standard_speech_v1.0 | — | cdli | 19.0 | low | |
| 291 | whisper-large-v3-turbo-turkish | openai/whisper-large-v3-turbo | selimc | 18.92 | medium | |
| 292 | moonshine-tiny-uk-gguf | UsefulSensors/moonshine-tiny-uk | handy-computer | 18.83 | low | |
| 293 | Kabyle_xlsr | — | Akashpb13 | 18.79 | medium | |
| 294 | whisper-large-v3-nepali | openai/whisper-large-v3 | kiranpantha | 18.73 | medium | |
| 295 | w2v-bert-waxal | sulaimank/w2v-bert-lingala-109h | sulaimank | 18.67 | medium | |
| 296 | wav2vec2-large-xls-r-1b-Swedish | facebook/wav2vec2-xls-r-1b | kingabzpro | 18.44 | medium | |
| 297 | whisper-small-bn | — | anuragshas | 18.22 | medium | |
| 298 | parakeet-ctc-110m-coreml | nvidia/parakeet-tdt_ctc-110m | FluidInference | 17.97 | medium | WER |
| 299 | whisper-small-uyghur-thugy20 | openai/whisper-small | ixxan | 17.94 | medium | |
| 300 | medasr-gguf | google/medasr | handy-computer | 17.88 | low | |
| 301 | wav2vec2-xls-r-300m-hebrew | facebook/wav2vec2-xls-r-300m | imvladikon | 17.73 | medium | |
| 302 | stt_uz_fastconformer_hybrid_large_pc | nvidia/stt_uz_fastconformer_hybrid_large_pc | nvidia | 17.52 | low | |
| 303 | xlsr-luganda-waxal-pseudo | — | sulaimank | 17.49 | medium | |
| 304 | w2v-bert-uk-v2.1 | facebook/w2v-bert-2.0 | Yehor | 17.34 | medium | |
| 305 | VibeVoice-ASR-HF | microsoft/VibeVoice-ASR-HF | microsoft | 17.2 | low | |
| 306 | parakeet-tdt-1.1b-gguf | nvidia/parakeet-tdt-1.1b | handy-computer | 17.18 | medium | % WER |
| 307 | parakeet-tdt-1.1b | nvidia/parakeet-tdt-1.1b | nvidia | 17.18 | medium | % WER |
| 308 | whisper-large-v3-atco2-asr | openai/whisper-large-v3 | jlvdoorn | 17.04 | medium | |
| 309 | canary-180m-flash-gguf | nvidia/canary-180m-flash | handy-computer | 16.92 | medium | % WER |
| 310 | canary-180m-flash | nvidia/canary-180m-flash | nvidia | 16.92 | medium | % WER |
| 311 | mms-300m-waxal-lug | facebook/mms-300m | waxal-benchmarking | 16.9 | medium | **Test WER** |
| 312 | mms-300m-waxal-lug | facebook/mms-300m | waxal-benchmarking | 16.9 | medium | **WER** |
| 313 | ne-asr | — | MWirelabs | 16.89 | low | |
| 314 | wav2vec2-xls-r-300m-en-atc-uwb-atcc-and-atcosim | — | Jzuluaga | 16.87 | medium | |
| 315 | asr-wav2vec2-commonvoice-14-en | — | speechbrain | 16.86 | low | |
| 316 | canary-qwen-2.5b | nvidia/canary-1b-flash | nvidia | 16.71 | medium | % WER |
| 317 | canary-qwen-2.5b-gguf | nvidia/canary-qwen-2.5b | handy-computer | 16.71 | medium | % WER |
| 318 | uzbek_stt_v1 | openai/whisper-medium | Kotib | 16.7 | low | |
| 319 | uzbek_stt_v1 | openai/whisper-medium | Koyiljon | 16.7 | low | |
| 320 | uzbek_stt_v1 | openai/whisper-medium | Uzbekswe | 16.7 | low | |
| 321 | stt_ar_fastconformer_hybrid_large_pcd_v1.0 | nvidia/stt_ar_fastconformer_hybrid_large_pcd_v1.0 | nvidia | 16.67 | low | |
| 322 | whisper-tiny_to_Chinese_accent_SP_PHaPS | openai/whisper-tiny | liuh6 | 16.58 | medium | |
| 323 | whisper-large-v3-turbo-tj | openai/whisper-large-v3-turbo | re-skill | 16.43 | medium | |
| 324 | kiswahili_sahihi_asr_adapted_2 | keystats/kiswahili_sahihi_asr | keystats | 16.23 | low | |
| 325 | whisper-base-ar-quran-mix-norm | openai/whisper-base | deepdml | 16.2 | medium | |
| 326 | wav2vec2-large-mms-1b-tatar | facebook/mms-1b-all | AigizK | 16.18 | medium | |
| 327 | whisper-small-sr | openai/whisper-small | istomin9192 | 16.14 | high | |
| 328 | surt-small-v3 | openai/whisper-small | surindersinghssj | 15.84 | medium | |
| 329 | w2vbert-lingala-waxal-punct-v2 | sulaimank/w2vbert-lingala-waxal | sulaimank | 15.72 | low | |
| 330 | whisper-base_ro-80mel | openai/whisper-base | IonGrozea | 15.58 | medium | WER |
| 331 | whisper-th-medium-timestamp | biodatlab/whisper-th-medium-combined | biodatlab | 15.57 | medium | |
| 332 | whisper-th-large-combined | — | biodatlab | 15.51 | medium | |
| 333 | DiarizationLM-8b-Fisher-v2 | — | google | 15.48 | low | |
| 334 | DiarizationLM-13b-Fisher-v1 | — | google | 15.48 | low | |
| 335 | DiarizationLM-8b-Fisher-v1 | — | google | 15.48 | low | |
| 336 | qwen3-asr-0.6b-hinglish | Qwen/Qwen3-ASR-0.6B | moorlee | 15.41 | low | |
| 337 | whisper-large-v3-turbo-armenian | openai/whisper-large-v3-turbo | Chillarmo | 15.31 | low | |
| 338 | whisper-large-v3-turbo-armenian | openai/whisper-large-v3-turbo | Chillarmo | 15.31 | medium | **Word Error Rate (WER)** |
| 339 | parakeet-tdt-0.6b-v3-int8 | sonic-speech/parakeet-tdt-0.6b-v3 | sonic-speech | 15.1 | medium | WER (TED-LIUM) |
| 340 | whisper-base_ro-80mel | openai/whisper-base | IonGrozea | 15.09 | medium | WER (normalized: locase, punctuation stripped) |
| 341 | za-anv-multilingual-whisper-v3-turbo | openai/whisper-large-v3-turbo | dsfsi-anv | 15.01 | medium | |
| 342 | whisper-large-v3-lingala-asr | openai/whisper-large-v3 | noirlab | 14.83 | medium | |
| 343 | whisper-large-finnish-v3 | — | Finnish-NLP | 14.81 | medium | |
| 344 | wav2vec2-xls-r-300m-cs-250 | facebook/wav2vec2-xls-r-300m | comodoro | 14.75 | medium | |
| 345 | NaijaVox-2.0 | openai/whisper-large-v3 | Axiveri | 14.7 | low | |
| 346 | whisper-small-pt-cv11-v5 | openai/whisper-small | jonatasgrosman | 14.68 | medium | |
| 347 | canary-1b-flash-gguf | nvidia/canary-1b-flash | handy-computer | 14.66 | medium | % WER |
| 348 | canary-1b-flash | nvidia/canary-1b-flash | nvidia | 14.66 | medium | % WER |
| 349 | canary-1b-flash-gguf | nvidia/canary-1b-flash | memoravox | 14.66 | medium | % WER |
| 350 | canary-1b-gguf | nvidia/canary-1b | handy-computer | 14.64 | medium | % WER |
| 351 | canary-1b | nvidia/canary-1b | nvidia | 14.64 | medium | % WER |
| 352 | wav2vec2-bert-CV16-en | ylacombe/w2v-bert-2.0 | hf-audio | 14.55 | medium | |
| 353 | moonshine-base-uk-gguf | UsefulSensors/moonshine-base-uk | handy-computer | 14.49 | low | |
| 354 | qwen3-asr-arabic-ksa | Qwen/Qwen3-ASR-1.7B | vadimbelsky | 14.41 | medium | **WER** |
| 355 | wav2vec2-base-japanese-asr | facebook/wav2vec2-base | TKU410410103 | 14.18 | medium | |
| 356 | whisper-large-fa-v1 | openai/whisper-large-v3-turbo | vhdm | 14.07 | medium | **Word Error Rate (WER)** |
| 357 | whisper-uz | jamshidahmadov/whisper-uz | jmshd | 14.01 | medium | |
| 358 | odia_whisper_small_v3.0 | — | Ranjit | 13.86 | medium | |
| 359 | whisper-medium-et | — | TalTechNLP | 13.8 | low | |
| 360 | canary-1b-v2 | nvidia/canary-1b-v2 | nvidia | 13.78 | low | |
| 361 | tadabur-Whisper-Small | openai/whisper-small | FaisaI | 13.78 | low | |
| 362 | sauti-asr | microsoft/paza-whisper-large-v3-turbo | msingiai | 13.72 | medium | WER |
| 363 | Nemotron-3.5-ASR-Streaming-0.6B-MLX-5bit | nvidia/nemotron-3.5-asr-streaming-0.6b | aufklarer | 13.66 | high | |
| 364 | whisper-large-v2-marathi | — | DrishtiSharma | 13.64 | medium | |
| 365 | whisper-base-es-mix-norm | openai/whisper-base | deepdml | 13.57 | medium | |
| 366 | IndicConformer-GGUF | ai4bharat/indicconformer_stt_hi_hybrid_rnnt_large | Singla0009 | 13.5 | medium | **IndicConformer GGUF (WER)** |
| 367 | whisper-small-ja | openai/whisper-small | vumichien | 13.47 | medium | |
| 368 | stt_kr_conformer_ctc_medium | — | SungBeom | 13.45 | medium | |
| 369 | whisper-large-v2-bg | — | anuragshas | 13.4 | medium | |
| 370 | Nemotron-3.5-ASR-Streaming-0.6B-MLX-8bit | nvidia/nemotron-3.5-asr-streaming-0.6b | aufklarer | 13.37 | high | |
| 371 | whisper-large-v3-turbo-russian | openai/whisper-large-v3-turbo | coriollon | 13.25 | medium | WER (aggregate) |
| 372 | moonshine-tiny-vi-gguf | UsefulSensors/moonshine-tiny-vi | handy-computer | 13.16 | low | |
| 373 | whisper-th-small-combined | openai/whisper-small | biodatlab | 13.14 | medium | |
| 374 | whisper-medium-hindi-hinglish-asr-fine-tuned | openai/whisper-medium | shujaAK | 13.01 | high | |
| 375 | sushrota-sanskrit-asr | — | prathoshap | 13.0 | low | |
| 376 | parakeet-tdt_ctc-1.1b-gguf | nvidia/parakeet-tdt_ctc-1.1b | handy-computer | 12.81 | medium | % WER |
| 377 | parakeet-tdt_ctc-1.1b | nvidia/parakeet-tdt_ctc-1.1b | nvidia | 12.81 | medium | % WER |
| 378 | w2vbert-luganda-waxal | sulaimank/w2v-bert-2.0-lg-CV-Fleurs-300 | sulaimank | 12.78 | medium | |
| 379 | wav2vec2-large-xlsr-german | — | maxidl | 12.77 | medium | |
| 380 | w2vbert-shona-waxal-punct-v2 | sulaimank/w2vbert-shona-waxal | sulaimank | 12.76 | low | |
| 381 | parakeet-tdt-0.6b-v3-ptBR-gguf | alexandreacff/parakeet-tdt-0.6b-v3-ptBR-plus | luizhrocha | 12.7 | low | |
| 382 | nb-wav2vec2-300m-nynorsk | — | NbAiLab | 12.22 | medium | |
| 383 | whisper-large-v3-turkish-general | openai/whisper-large-v3 | turkmedstt | 12.13 | low | |
| 384 | wav2vec2-base-german-cv9 | facebook/wav2vec2-base | oliverguhr | 12.09 | medium | |
| 385 | nemotron-ocr-v1 | nvidia/nemotron-ocr-v1 | nvidia | 12.03 | medium | Bag-of-word Error Rate |
| 386 | parakeet-tdt-0.6b-v3-gguf | nvidia/parakeet-tdt-0.6b-v3 | handy-computer | 11.97 | medium | **Average WER ↓** |
| 387 | parakeet-tdt-0.6b-v3 | nvidia/parakeet-tdt-0.6b-v3 | nvidia | 11.97 | medium | **Average WER ↓** |
| 388 | kazakh-whisper-large-v3-turbo | openai/whisper-large-v3-turbo | shyngys879 | 11.8 | low | |
| 389 | hviske-v5 | — | syvai | 11.8 | low | |
| 390 | romanian-wav2vec2 | facebook/wav2vec2-xls-r-300m | gigant | 11.74 | medium | |
| 391 | Nemotron-Speech-Streaming-0.6B-CoreML-INT8 | nvidia/nemotron-speech-streaming-en-0.6b | aufklarer | 11.73 | high | |
| 392 | wav2vec2-large-xlsr-53-greek | — | jonatasgrosman | 11.62 | high | |
| 393 | stt_ar_fastconformer_hybrid_large_pc_v1.0 | nvidia/stt_ar_fastconformer_hybrid_large_pc_v1.0 | nvidia | 11.46 | low | |
| 394 | moonshine-tiny-de-fidoriel-GGUF | fidoriel/moonshine-tiny-de | cstr | 11.4 | medium | |
| 395 | nb-wav2vec2-1b-nynorsk | — | NbAiLab | 11.32 | medium | |
| 396 | Zipformer-cr-ctc-transducer-XL-290M | — | soundsgoodai | 11.3 | low | |
| 397 | Shrutimala_Bangla_ASR | facebook/w2v-bert-2.0 | sazzadul | 11.26 | medium | |
| 398 | whisper-large-v2-bn | — | anuragshas | 11.11 | medium | |
| 399 | omi-med-stt-v1 | nvidia/parakeet-tdt-0.6b-v2 | omi-health | 11.1 | low | |
| 400 | omi-med-stt-v1-mlx-q8 | omi-health/omi-med-stt-v1 | omi-health | 11.1 | low | |
| 401 | kiswahili_sahihi_asr_adapted_2 | keystats/kiswahili_sahihi_asr | keystats | 11.09 | medium | **Final Validation WER** |
| 402 | whisper-large-v3-turbo-greek-greece | openai/whisper-large-v3-turbo | sam8000 | 11.06 | medium | |
| 403 | cohere-transcribe-arabic-07-2026-gguf | CohereLabs/cohere-transcribe-arabic-07-2026 | handy-computer | 11.02 | low | |
| 404 | Qwen3-ASR-1.7B-nvfp4 | Qwen/Qwen3-ASR-1.7B | vrfai | 10.73 | high | |
| 405 | parakeet-rnnt-110m-da-dk | nvidia/parakeet-rnnt-110m-da-dk | nvidia | 10.7 | low | |
| 406 | wav2vec2-xls-r-300m-cv8-turkish | — | mpoyraz | 10.61 | low | |
| 407 | wav2vec2-large-xlsr-53-greek | — | lighteternal | 10.5 | medium | WER on CommonVoice Test (%) * |
| 408 | wav2vec2-large-960h-lv60-self-en-atc-uwb-atcc-and-atcosim | facebook/wav2vec2-large-960h-lv60-self | Jzuluaga | 10.49 | medium | |
| 409 | wav2vec2-xls-r-parlaspeech-hr | — | classla | 10.46 | low | |
| 410 | parakeet-tdt-0.6b-v3-fr-tv-media | nvidia/parakeet-tdt-0.6b-v3 | Archime | 10.42 | low | |
| 411 | wav2vec2-large-xlsr-persian-v3 | — | m3hrdadfi | 10.36 | medium | |
| 412 | hviske-v5.3 | syvai/hviske-v5.1 | syvai | 10.26 | low | |
| 413 | higgs-audio-v3-8b-stt-v2 | bosonai/higgs-audio-v3-8b | bosonai | 10.14 | low | |
| 414 | Cohere-Speech-Tashkeel-2B | CohereLabs/cohere-transcribe-arabic-07-2026 | NAMAA-Space | 10.13 | low | |
| 415 | Cohere-Speech-Tashkeel-2B | CohereLabs/cohere-transcribe-arabic-07-2026 | NAMAA-Space | 10.13 | medium | **WER — undiacritized** |
| 416 | higgs-audio-v3-stt-v2 | bosonai/higgs-audio-v3-stt | bosonai | 10.03 | low | |
| 417 | whisper-large-v3-turbo-bulgarian-bulgaria | openai/whisper-large-v3-turbo | sam8000 | 9.97 | medium | |
| 418 | whisper-large-v3-russian | antony66/whisper-large-v3-russian | antony66 | 9.84 | medium | |
| 419 | hindi-tts-190M | — | 5ivatej | 9.8 | low | |
| 420 | whisper-small-singlish-122k | openai/whisper-small | jensenlwt | 9.69 | medium | |
| 421 | moonshine-base-vi-gguf | UsefulSensors/moonshine-base-vi | handy-computer | 9.67 | low | |
| 422 | whisper-large-v3-it-cv16 | openai/whisper-large | fsicoli | 9.63 | medium | |
| 423 | parakeet-tdt-0.6b-portuguese | nvidia/parakeet-tdt-0.6b-v3 | yuriyvnv | 9.62 | low | |
| 424 | w2vbert-lingala-waxal | sulaimank/w2v-bert-lingala-109h | sulaimank | 9.61 | medium | |
| 425 | Nemotron-3.5-ASR-Streaming-0.6B-CoreML-INT8 | nvidia/nemotron-3.5-asr-streaming-0.6b | aufklarer | 9.59 | low | |
| 426 | Nemotron-3.5-ASR-Streaming-0.6B-CoreML-INT8 | nvidia/nemotron-3.5-asr-streaming-0.6b | playstonex | 9.59 | low | |
| 427 | wav2vec2-large-xlsr-53-german-cv9 | — | oliverguhr | 9.48 | medium | |
| 428 | whisper-small-ko | — | SungBeom | 9.48 | low | |
| 429 | whisper-accent-medium.en | openai/whisper-medium.en | mavleo96 | 9.39 | medium | |
| 430 | whisper-large-v3-spanish | openai/whisper-large-v3 | Berly00 | 9.36 | medium | |
| 431 | parakeet-ctc-0.6b-Vietnamese | nvidia/parakeet-ctc-0.6b-Vietnamese | nvidia | 9.3 | low | |
| 432 | parakeet_realtime_eou_120m-v1 | nvidia/parakeet_realtime_eou_120m-v1 | nvidia | 9.3 | medium | WER (%) |
| 433 | Voxtral-Mini-3B-2507-AWQ4-ARG | mistralai/Voxtral-Mini-3B-2507 | juanginer | 9.24 | medium | WER nl |
| 434 | Voxtral-Mini-3B-2507-AWQ4-GEN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 9.24 | medium | WER nl |
| 435 | Voxtral-Mini-3B-2507-AWQ4-EN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 9.24 | medium | WER nl |
| 436 | wav2vec2-large-slavic-parlaspeech-hr | — | classla | 9.21 | low | |
| 437 | whisper-small_ro-80mel | openai/whisper-small | IonGrozea | 9.21 | medium | WER |
| 438 | shook-medium-amharic-2k | — | b1n1yam | 9.11 | medium | |
| 439 | whisper-medium-jp | — | vumichien | 9.04 | medium | |
| 440 | whisper-large-v3-turbo-russian | openai/whisper-large-v3-turbo | coriollon | 8.93 | medium | |
| 441 | whisper-small_ro-80mel | openai/whisper-small | IonGrozea | 8.8 | medium | WER (normalized: locase, punctuation stripped) |
| 442 | xls-r-300m-danish-mc-v2 | MediaCatch/xls-r-300m-danish-mc | MediaCatch | 8.75 | medium | |
| 443 | whisper-persian-v4 | openai/whisper-large-v3 | nezamisafa | 8.73 | medium | |
| 444 | Pathumma-whisper-th-large-v3 | openai/whisper-large-v3 | nectec | 8.68 | high | |
| 445 | wav2vec2-xls-r-300m-cv7-turkish | — | mpoyraz | 8.62 | low | |
| 446 | Shenava-Koochik-0.9 | nvidia/stt_en_fastconformer_hybrid_large_streaming_multi | Reza2kn | 8.57 | low | |
| 447 | MOSS-Transcribe-Diarize-0.9B-MLX-5bit | OpenMOSS-Team/MOSS-Transcribe-Diarize | aufklarer | 8.55 | low | |
| 448 | kid-whisper-small-myst | — | aadel4 | 8.51 | medium | |
| 449 | wav2vec2-large-xlsr-53-spanish-with-lm | — | patrickvonplaten | 8.44 | high | |
| 450 | gigaam-v3-ctc-gguf | ai-sage/GigaAM-v3 | handy-computer | 8.42 | low | |
| 451 | distil-small.en | distil-whisper/distil-small.en | distil-whisper | 8.4 | low | |
| 452 | distil-medium.en | — | distil-whisper | 8.4 | low | |
| 453 | distil-large-v2 | distil-whisper/distil-large-v3 | distil-whisper | 8.4 | low | |
| 454 | MOSS-Transcribe-preview-2B | OpenMOSS-Team/MOSS-Transcribe-preview-2B | OpenMOSS-Team | 8.37 | low | |
| 455 | parakeet-realtime-eou-120m-coreml | nvidia/parakeet_realtime_eou_120m-v1 | FluidInference | 8.29 | low | |
| 456 | wav2vec2-xlsr-300m-finnish-lm | — | Finnish-NLP | 8.16 | high | |
| 457 | whisper-large-v3-turbo-fr-quebecois | openai/whisper-large-v3-turbo | ele-sage | 8.1 | low | |
| 458 | gigaam-v3-rnnt-gguf | ai-sage/GigaAM-v3 | handy-computer | 8.08 | low | |
| 459 | Voxtral-Mini-3B-2507-AWQ4-ARG | mistralai/Voxtral-Mini-3B-2507 | juanginer | 8.01 | medium | WER fr |
| 460 | Voxtral-Mini-3B-2507-AWQ4-GEN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 8.01 | medium | WER fr |
| 461 | Voxtral-Mini-3B-2507-AWQ4-EN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 8.01 | medium | WER fr |
| 462 | Freya-TTS | — | freyavoice | 8.0 | medium | |
| 463 | nemotron-3.5-asr-streaming-0.6b-gguf | nvidia/nemotron-3.5-asr-streaming-0.6b | handy-computer | 7.97 | low | |
| 464 | nemotron-3.5-asr-streaming-0.6b-gguf | nvidia/nemotron-3.5-asr-streaming-0.6b | memoravox | 7.97 | low | |
| 465 | xls-r-300m-et | — | TalTechNLP | 7.9 | low | |
| 466 | wav2vec2-base-word-by-word-quran-asr | facebook/wav2vec2-base | HamzaSidhu786 | 7.9 | medium | |
| 467 | whisper-large-v3-czech-cv13 | openai/whisper-large-v3 | mikr | 7.89 | medium | |
| 468 | Nos_ASR-wav2vec2-xls-r-300m-gl | facebook/wav2vec2-xls-r-300m | proxectonos | 7.85 | low | |
| 469 | ATC-whisper-turbo-v1 | openai/whisper-large-v3-turbo | SingularityUS | 7.83 | low | |
| 470 | whisper-large-v3-turbo-verbatim-1 | openai/whisper-large-v3-turbo | JacobLinCool | 7.73 | medium | |
| 471 | Qwen3-ASR-1.7B-fp8 | Qwen/Qwen3-ASR-1.7B | vrfai | 7.6 | high | |
| 472 | whisper-large-v3-pt-cv19-fleurs | openai/whisper-large-v3 | fsicoli | 7.56 | medium | |
| 473 | whisper-tiny-gguf | openai/whisper-tiny | handy-computer | 7.54 | low | |
| 474 | LFM2.5-Audio-1.5B | LiquidAI/LFM2-1.2B | LiquidAI | 7.53 | high | Open |
| 475 | LFM2.5-Audio-1.5B-GGUF | LiquidAI/LFM2-1.2B | Mungert | 7.53 | high | Open |
| 476 | Shenava-Koochik-v1.0 | nvidia/stt_fa_fastconformer_hybrid_large | Reza2kn | 7.49 | low | |
| 477 | whisper-hu-large-v3-turbo-finetuned | openai/whisper-large-v3-turbo | sarpba | 7.48 | medium | |
| 478 | whisper-th-medium-combined | openai/whisper-medium | biodatlab | 7.42 | medium | |
| 479 | distil-large-v3.5 | distil-whisper/distil-large-v3.5 | distil-whisper | 7.3 | low | |
| 480 | wav2vec2-2-bart-base | — | patrickvonplaten | 7.28 | medium | |
| 481 | whisper-large-v2-pl-v2 | — | bardsai | 7.28 | medium | |
| 482 | Voxtral-Mini-3B-2507-AWQ4-ARG | mistralai/Voxtral-Mini-3B-2507 | juanginer | 7.16 | medium | WER pt |
| 483 | Voxtral-Mini-3B-2507-AWQ4-GEN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 7.16 | medium | WER pt |
| 484 | Voxtral-Mini-3B-2507-AWQ4-EN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 7.16 | medium | WER pt |
| 485 | Voxtral-Mini-3B-2507-AWQ4-ARG | mistralai/Voxtral-Mini-3B-2507 | juanginer | 7.13 | medium | WER avg-6 |
| 486 | Voxtral-Mini-3B-2507-AWQ4-GEN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 7.13 | medium | WER avg-6 |
| 487 | Voxtral-Mini-3B-2507-AWQ4-EN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 7.13 | medium | WER avg-6 |
| 488 | g-omnivoice | k2-fsa/OmniVoice | g-group-ai-lab | 7.12 | low | |
| 489 | whisper-tiny-ar-quran | — | tarteel-ai | 7.05 | medium | |
| 490 | nb-wav2vec2-300m-bokmaal | — | NbAiLab | 7.03 | medium | |
| 491 | VoiceTut-TTS | k2-fsa/OmniVoice | mohammedaly22 | 7.0 | medium | WER — English |
| 492 | nemotron-speech-streaming-en-0.6b | nvidia/nemotron-speech-streaming-en-0.6b | nvidia | 6.93 | medium | **WER (%)** |
| 493 | nemotron-speech-streaming-en-0.6b-gguf | nvidia/nemotron-speech-streaming-en-0.6b | handy-computer | 6.93 | medium | **WER (%)** |
| 494 | nemotron-3.5-asr-streaming-0.6b-GGUF | nvidia/nemotron-3.5-asr-streaming-0.6b | cstr | 6.93 | high | |
| 495 | nemotron-3.5-asr-streaming-GGUF | nvidia/nemotron-3.5-asr-streaming-0.6b | cstr | 6.93 | high | |
| 496 | nemotron-speech-streaming-en-0.6b-gguf | nvidia/nemotron-speech-streaming-en-0.6b | memoravox | 6.93 | medium | **WER (%)** |
| 497 | wav2vec2-large-xlsr-catala | — | softcatala | 6.92 | low | |
| 498 | moonshine-base-de-fidoriel-GGUF | fidoriel/moonshine-base-de | cstr | 6.9 | medium | |
| 499 | moonshine-base-de | UsefulSensors/moonshine-base | fidoriel | 6.9 | medium | |
| 500 | wavlm-libri-clean-100h-base-plus | — | patrickvonplaten | 6.83 | medium | |
| 501 | asr-conformer-loquacious | — | speechbrain | 6.8 | low | |
| 502 | wavlm-libri-clean-100h-base | — | patrickvonplaten | 6.75 | medium | |
| 503 | Qwen3-ASR-1.7B-NL | Qwen/Qwen3-ASR-1.7B | yuriyvnv | 6.7 | low | Zero-shot WER | CER (zero-shot → fine-tuned) |
| 504 | f5-tts-quebec-french-finetune | RASPIAUDIO/F5-French-MixedSpeakers-reduced | tontate | 6.7 | medium | WER ↓ (faster-whisper large-v3) |
| 505 | wav2vec2-large-danish-npsc-nst | chcaa/xls-r-300m-danish | NbAiLab | 6.69 | medium | |
| 506 | CrisperWhisper | openai/whisper-large-v3 | nyralabs | 6.66 | medium | **Average WER** |
| 507 | CrisperWhisper | openai/whisper-large-v3 | unsloth | 6.66 | medium | **Average WER** |
| 508 | Voxtral-Mini-3B-2507-AWQ4-ARG | mistralai/Voxtral-Mini-3B-2507 | juanginer | 6.65 | medium | WER en (no MLS) |
| 509 | Voxtral-Mini-3B-2507-AWQ4-GEN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 6.65 | medium | WER en (no MLS) |
| 510 | Voxtral-Mini-3B-2507-AWQ4-EN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 6.65 | medium | WER en (no MLS) |
| 511 | whisper-medium_ro-80mel | openai/whisper-medium | IonGrozea | 6.63 | medium | WER |
| 512 | whisper-medium-portuguese | — | pierreguillou | 6.6 | medium | |
| 513 | w2v-bert-uk | facebook/w2v-bert-2.0 | Yehor | 6.6 | medium | |
| 514 | whisper-th-large-v3-combined | openai/whisper-large-v3 | biodatlab | 6.59 | medium | |
| 515 | Voxtral-Mini-3B-2507-AWQ4-ARG | mistralai/Voxtral-Mini-3B-2507 | juanginer | 6.5 | medium | WER it |
| 516 | Voxtral-Mini-3B-2507-AWQ4-GEN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 6.5 | medium | WER it |
| 517 | Voxtral-Mini-3B-2507-AWQ4-EN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 6.5 | medium | WER it |
| 518 | Fasih-TTS-V1 | coqui/XTTS-v2 | NightPrince | 6.5 | low | |
| 519 | parakeet-tdt-0.6b-v3 | nvidia/parakeet-tdt-0.6b-v3 | nvidia | 6.34 | low | |
| 520 | canary-1b-v2-GGUF | nvidia/canary-1b-v2 | cstr | 6.34 | medium | Open ASR WER (avg) |
| 521 | nb-wav2vec2-1b-bokmaal | — | NbAiLab | 6.33 | medium | |
| 522 | Borealis | — | Vikhrmodels | 6.33 | low | |
| 523 | Qwen3-ASR-0.6B-hf | Qwen/Qwen3-ASR-0.6B-hf | Qwen | 6.31 | high | |
| 524 | Qwen3-ASR-0.6B-hf | Qwen/Qwen3-ASR-0.6B | bezzam | 6.31 | high | |
| 525 | whisper-medium_ro-80mel | openai/whisper-medium | IonGrozea | 6.3 | medium | WER (normalized: locase, punctuation stripped) |
| 526 | whisper-large-id | — | cahya | 6.25 | medium | |
| 527 | higgs-audio-v3-8b-stt | bosonai/higgs-audio-v3-8b | bosonai | 6.23 | low | |
| 528 | Phi-4-multimodal-instruct | microsoft/Phi-4-multimodal-instruct | microsoft | 6.14 | medium | |
| 529 | Phi-4-multimodal-instruct | microsoft/Phi-4-multimodal-instruct | Lexius | 6.14 | medium | |
| 530 | whisper-small-ckb | openai/whisper-small | razhan | 6.13 | medium | |
| 531 | thewhisper-large-v3-turbo | openai/whisper-large-v3-turbo | TheStageAI | 6.12 | medium | **Mean WER** |
| 532 | Voxtral-Mini-3B-2507-AWQ4-ARG | mistralai/Voxtral-Mini-3B-2507 | juanginer | 6.06 | medium | WER de |
| 533 | Voxtral-Mini-3B-2507-AWQ4-GEN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 6.06 | medium | WER de |
| 534 | Voxtral-Mini-3B-2507-AWQ4-EN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 6.06 | medium | WER de |
| 535 | whisper-small-id | — | cahya | 6.06 | medium | |
| 536 | parakeet-tdt-0.6b-v2 | nvidia/parakeet-tdt-0.6b-v2 | nvidia | 6.05 | high | |
| 537 | parakeet-tdt-0.6b-v2-gguf | nvidia/parakeet-tdt-0.6b-v2 | handy-computer | 6.05 | high | |
| 538 | wav2vec2-large-100k-voxpopuli-catala | — | ccoreilly | 5.98 | low | |
| 539 | Nemotron-3-Nano-Omni-30B-A3B-Reasoning-NVFP4 | nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | nvidia | 5.95 | medium | HF-ASR (WER↓) |
| 540 | Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8 | nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | nvidia | 5.95 | medium | HF-ASR (WER↓) |
| 541 | Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | nvidia | 5.95 | medium | HF-ASR (WER↓) |
| 542 | NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-GGUF | nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | unsloth | 5.95 | medium | HF-ASR (WER↓) |
| 543 | NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning | nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | unsloth | 5.95 | medium | HF-ASR (WER↓) |
| 544 | moonshine_tiny_pt_v08 | aomocelin/moonshine_tiny_pt_v05 | aomocelin | 5.86 | medium | |
| 545 | Voxtral-Mini-3B-2507-AWQ4-ARG | mistralai/Voxtral-Mini-3B-2507 | juanginer | 5.82 | medium | WER es |
| 546 | Voxtral-Mini-3B-2507-AWQ4-GEN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 5.82 | medium | WER es |
| 547 | Voxtral-Mini-3B-2507-AWQ4-EN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 5.82 | medium | WER es |
| 548 | whisper-tiny.en-gguf | openai/whisper-tiny.en | handy-computer | 5.77 | low | |
| 549 | whisper-base-ar-quran | openai/whisper-base | tarteel-ai | 5.75 | medium | |
| 550 | wav2vec2-xlsr-1b-finnish-lm-v2 | — | Finnish-NLP | 5.65 | high | |
| 551 | Qwen3-ASR-1.7B-hf | Qwen/Qwen3-ASR-1.7B-hf | Qwen | 5.59 | high | |
| 552 | Qwen3-ASR-1.7B-hf | Qwen/Qwen3-ASR-1.7B | bezzam | 5.59 | high | |
| 553 | wav2vec2-large-slavic-parlaspeech-hr-lm | — | classla | 5.56 | low | |
| 554 | Zipformer-cr-ctc-transducer-XL-290M | — | soundsgoodai | 5.56 | medium | Average WER |
| 555 | Whisper-Large-v3-Turbo-CoreML | openai/whisper-large-v3-turbo | aufklarer | 5.55 | low | |
| 556 | granite-4.0-1b-speech | ibm-granite/granite-4.0-1b-base | ibm-granite | 5.52 | high | |
| 557 | granite-4.0-1b-speech-ONNX | ibm-granite/granite-4.0-1b-speech | onnx-community | 5.52 | low | |
| 558 | gigaam-v3-e2e-ctc-gguf | ai-sage/GigaAM-v3 | handy-computer | 5.5 | low | |
| 559 | gigaam-v3-e2e-ctc-gguf | ai-sage/GigaAM-v3 | memoravox | 5.5 | low | |
| 560 | cohere-transcribe-03-2026-GGUF | CohereLabs/cohere-transcribe-03-2026 | cstr | 5.42 | medium | |
| 561 | gigaam-v3-e2e-rnnt-gguf | ai-sage/GigaAM-v3 | handy-computer | 5.35 | low | |
| 562 | gigaam-v3-e2e-rnnt-gguf | ai-sage/GigaAM-v3 | memoravox | 5.35 | low | |
| 563 | granite-speech-4.1-2b-plus | ibm-granite/granite-4.0-1b-base | ibm-granite | 5.33 | high | |
| 564 | whisper-base-gguf | openai/whisper-base | handy-computer | 5.1 | low | |
| 565 | whisper-base-gguf | openai/whisper-base | memoravox | 5.1 | low | |
| 566 | cohere-transcribe-arabic-07-2026-MLX-Mixed-2bit3bit4bit | CohereLabs/cohere-transcribe-arabic-07-2026 | MarkChen1214 | 5.09 | low | |
| 567 | gigaam-multilingual-mlx | ai-sage/GigaAM-Multilingual | ai-babai | 5.07 | low | |
| 568 | gigaam-multilingual-mlx-int4-g64 | ai-babai/gigaam-multilingual-mlx | ai-babai | 5.07 | low | |
| 569 | gigaam-multilingual-mlx-int8-g64 | ai-babai/gigaam-multilingual-mlx | ai-babai | 5.07 | low | |
| 570 | gigaam-multilingual-mlx-int6-g64 | ai-babai/gigaam-multilingual-mlx | ai-babai | 5.07 | low | |
| 571 | best_for_now | openai/whisper-small | MosesJoshuaCoker | 5.05 | medium | Word Error Rate (WER) |
| 572 | wavlm-libri-clean-100h-large | — | patrickvonplaten | 4.91 | medium | |
| 573 | whisper-tiny-yoruba-v2 | openai/whisper-tiny | EYEDOL | 4.91 | medium | |
| 574 | MOSS-TTSD-v1.0 | — | OpenMOSS-Team | 4.85 | high | |
| 575 | whisper-large-v3-lv-late-cv19 | openai/whisper-large-v3 | AiLab-IMCS-UL | 4.8 | low | |
| 576 | wav2vec2-large-xlsr-korean | — | kresnik | 4.74 | medium | |
| 577 | Voxtral-Mini-3B-2507-AWQ4-ARG | mistralai/Voxtral-Mini-3B-2507 | juanginer | 4.69 | medium | WER es-AR |
| 578 | Voxtral-Mini-3B-2507-AWQ4-GEN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 4.69 | medium | WER es-AR |
| 579 | Voxtral-Mini-3B-2507-AWQ4-EN | mistralai/Voxtral-Mini-3B-2507 | juanginer | 4.69 | medium | WER es-AR |
| 580 | moonshine-tiny-gguf | UsefulSensors/moonshine-tiny | handy-computer | 4.58 | low | |
| 581 | BanglaASR | — | bangla-speech-processing | 4.58 | medium | |
| 582 | whisper-medium-hindi-hinglish-asr-fine-tuned | openai/whisper-medium | shujaAK | 4.56 | medium | Word Error Rate (WER) |
| 583 | whisper-medium-hindi-hinglish-asr-fine-tuned | openai/whisper-medium | shujaAK | 4.56 | medium | WER |
| 584 | moonshine-tiny-GGUF | UsefulSensors/moonshine-tiny | cstr | 4.55 | medium | |
| 585 | moonshine-streaming-tiny-gguf | UsefulSensors/moonshine-streaming-tiny | handy-computer | 4.53 | low | |
| 586 | whisper-large-v3-high-mixed-nl | openai/whisper-large-v3 | yuriyvnv | 4.43 | medium | **Test WER (Common Voice)** |
| 587 | dots.tts-mf | rednote-hilab/dots.tts-soar | dots-studio | 4.4 | low | |
| 588 | qwen3-asr-0.6b-coreml | Qwen/Qwen3-ASR-0.6B | FluidInference | 4.4 | low | |
| 589 | Inflect-Nano-v2 | owensong/Inflect-Nano-v1 | owensong | 4.21 | low | |
| 590 | Inflect-Nano-v2-GGUF | owensong/Inflect-Nano-v2 | remixerdec | 4.21 | low | |
| 591 | whisper-base.en-gguf | openai/whisper-base.en | handy-computer | 4.14 | low | |
| 592 | wav2vec2-large-uralic-voxpopuli-v2-finnish | — | Finnish-NLP | 4.13 | high | |
| 593 | fastconformer-quran | nvidia/stt_ar_fastconformer_hybrid_large_pcd_v1.0 | Muno459 | 4.13 | low | |
| 594 | Inflect-Micro-v2 | owensong/Inflect-Micro-v2 | owensong | 3.99 | low | |
| 595 | Inflect-Micro-v2-GGUF | owensong/Inflect-Micro-v2 | remixerdec | 3.99 | low | |
| 596 | WhisperLv3-FT | openai/whisper-large-v3 | inesc-id | 3.97 | medium | |
| 597 | whisper-medium-id | openai/whisper-medium | cahya | 3.83 | medium | |
| 598 | whisper-large-v3-high-mixed-nl | openai/whisper-large-v3 | yuriyvnv | 3.57 | medium | **Validation WER** |
| 599 | whisper-large-v3-high-mixed-nl | openai/whisper-large-v3 | yuriyvnv | 3.56 | low | |
| 600 | whisper-large-v3-turbo-hr-parla | openai/whisper-large-v3 | GoranS | 3.52 | medium | |
| 601 | wav2vec2-1b-npsc-nst-bokmaal-repaired | facebook/wav2vec2-xls-r-1b | NbAiLab | 3.49 | medium | |
| 602 | whisper-tiny-ar-quran-mix-norm | openai/whisper-tiny | deepdml | 3.39 | medium | |
| 603 | whisper-small-gguf | openai/whisper-small | handy-computer | 3.33 | low | |
| 604 | moonshine-base-gguf | UsefulSensors/moonshine-base | handy-computer | 3.28 | low | |
| 605 | SenseVoiceSmall-gguf | FunAudioLLM/SenseVoiceSmall | handy-computer | 3.13 | low | |
| 606 | whisper-small-finetuned-medical3 | openai/whisper-small | Johnyquest7 | 3.12 | medium | |
| 607 | Nemotron-3-Nano-Omni-30B-A3B-Reasoning-NVFP4 | nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | nvidia | 3.11 | medium | Tedium Long (WER↓) |
| 608 | Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8 | nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | nvidia | 3.11 | medium | Tedium Long (WER↓) |
| 609 | Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | nvidia | 3.11 | medium | Tedium Long (WER↓) |
| 610 | NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-GGUF | nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | unsloth | 3.11 | medium | Tedium Long (WER↓) |
| 611 | NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning | nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | unsloth | 3.11 | medium | Tedium Long (WER↓) |
| 612 | parakeet-tdt-0.6b-v3-coreml | nvidia/parakeet-tdt-0.6b-v3 | mweinbach1 | 3.1 | low | |
| 613 | whisper-small.en-gguf | openai/whisper-small.en | handy-computer | 3.09 | low | |
| 614 | asr-crdnn-rnnlm-librispeech | — | speechbrain | 3.09 | low | |
| 615 | moonshine_tiny_pt_v06 | aomocelin/moonshine_tiny_pt_v05 | aomocelin | 3.0 | medium | |
| 616 | parakeet-tdt-ctc-110m-coreml | nvidia/parakeet-tdt_ctc-110m | FluidInference | 3.0 | low | |
| 617 | w2vbert-shona-waxal | sulaimank/W2V2_Bert_Afrivoice_FLEURS_Shona_100hr_v1 | sulaimank | 2.96 | medium | |
| 618 | mtl-mimic-voicebank | — | speechbrain | 2.89 | low | |
| 619 | dots.tts-soar | rednote-hilab/dots.tts-base | dots-studio | 2.8 | low | |
| 620 | dots.tts-base | rednote-hilab/dots.tts-base | dots-studio | 2.8 | low | |
| 621 | ZONOS2-GGUF | Zyphra/ZONOS2 | Zyphra | 2.79 | low | |
| 622 | whisper-medium.en-gguf | openai/whisper-medium.en | handy-computer | 2.74 | low | |
| 623 | whisper-large-gguf | openai/whisper-large | handy-computer | 2.72 | low | |
| 624 | asr-streaming-conformer-librispeech | — | speechbrain | 2.72 | low | |
| 625 | Voila-Tokenizer | — | maitrix-org | 2.7 | low | |
| 626 | whisper-large-v2-gguf | openai/whisper-large-v2 | handy-computer | 2.68 | low | |
| 627 | whisper-medium-gguf | openai/whisper-medium | handy-computer | 2.63 | low | |
| 628 | whisper-medium-gguf | openai/whisper-medium | memoravox | 2.63 | low | |
| 629 | cohere-transcribe-03-2026-CoreML-6bit | CohereLabs/cohere-transcribe-03-2026 | BarathwajAnandan | 2.58 | low | |
| 630 | CosyVoice2-0.5B | FunAudioLLM/CosyVoice2-0.5B | FunAudioLLM | 2.57 | high | Open-Source |
| 631 | whisper-small-ar-quran-mix-norm | openai/whisper-small | deepdml | 2.54 | medium | |
| 632 | moonshine-streaming-small-gguf | UsefulSensors/moonshine-streaming-small | handy-computer | 2.53 | low | |
| 633 | whisperkit-coreml_01-30-24 | — | argmaxinc | 2.44 | low | |
| 634 | parakeet-tdt_ctc-110m-gguf | nvidia/parakeet-tdt_ctc-110m | handy-computer | 2.43 | low | |
| 635 | VibeVoice-Realtime-0.5B | Qwen/Qwen2.5-0.5B | microsoft | 2.4 | low | |
| 636 | VibeVoice-Realtime-0.5B-ONNX | microsoft/VibeVoice-Realtime-0.5B | elbruno | 2.4 | low | |
| 637 | parakeet-unified-en-0.6b-coreml | nvidia/parakeet-tdt-0.6b-v2 | FluidInference | 2.37 | low | |
| 638 | nemotron-speech-streaming-en-0.6b-gguf | nvidia/nemotron-speech-streaming-en-0.6b | handy-computer | 2.31 | low | |
| 639 | nemotron-speech-streaming-en-0.6b-gguf | nvidia/nemotron-speech-streaming-en-0.6b | memoravox | 2.31 | low | |
| 640 | Breeze-ASR-25-gguf | MediaTek-Research/Breeze-ASR-25 | handy-computer | 2.29 | low | |
| 641 | asr-transformer-transformerlm-librispeech | — | speechbrain | 2.27 | low | |
| 642 | Fun-CosyVoice3-0.5B-2512 | FunAudioLLM/Fun-CosyVoice3-0.5B-2512 | FunAudioLLM | 2.24 | high | Open-Source |
| 643 | multitalker-parakeet-streaming-0.6b-v1-gguf | nvidia/multitalker-parakeet-streaming-0.6b-v1 | handy-computer | 2.19 | low | |
| 644 | moonshine-streaming-medium-gguf | UsefulSensors/moonshine-streaming-medium | handy-computer | 2.16 | low | |
| 645 | CosyVoice-300M-SFT | — | FunAudioLLM | 2.14 | low | Open-Source |
| 646 | CosyVoice-300M | FunAudioLLM/CosyVoice-300M | FunAudioLLM | 2.14 | low | Open-Source |
| 647 | CosyVoice-300M-Instruct | FunAudioLLM/CosyVoice-300M | FunAudioLLM | 2.14 | low | Open-Source |
| 648 | Raon-OpenTTS-1B | — | KRAFTON | 2.14 | low | |
| 649 | Qwen3-ASR-0.6B-gguf | Qwen/Qwen3-ASR-0.6B | handy-computer | 2.11 | low | |
| 650 | Voxtral-Mini-4B-Realtime-2602-gguf | mistralai/Voxtral-Mini-4B-Realtime-2602 | handy-computer | 2.08 | low | |
| 651 | moss-transcribe-diarize-gguf | OpenMOSS-Team/MOSS-Transcribe-Diarize | handy-computer | 2.08 | low | |
| 652 | Voxtral-Mini-4B-Realtime-2602-gguf | mistralai/Voxtral-Mini-4B-Realtime-2602 | memoravox | 2.08 | low | |
| 653 | asr-conformer-transformerlm-librispeech | — | speechbrain | 2.01 | low | |
| 654 | whisper-large-v3-turbo-gguf | openai/whisper-large-v3-turbo | handy-computer | 2.0 | low | |
| 655 | parakeet-tdt-0.6b-v3-gguf | nvidia/parakeet-tdt-0.6b-v3 | handy-computer | 1.95 | low | |
| 656 | canary-180m-flash-gguf | nvidia/canary-180m-flash | handy-computer | 1.94 | low | |
| 657 | canary-1b-v2-gguf | nvidia/canary-1b-v2 | handy-computer | 1.92 | low | |
| 658 | canary-1b-v2-gguf | nvidia/canary-1b-v2 | memoravox | 1.92 | low | |
| 659 | cohere-transcribe-arabic-07-2026-int4 | CohereLabs/cohere-transcribe-arabic-07-2026 | NAMAA-Space | 1.89 | low | |
| 660 | Voxtral-Mini-3B-2507-gguf | mistralai/Voxtral-Mini-3B-2507 | handy-computer | 1.88 | low | |
| 661 | parakeet-ctc-0.6b-gguf | nvidia/parakeet-ctc-0.6b | handy-computer | 1.87 | low | |
| 662 | parakeet-tdt_ctc-1.1b-gguf | nvidia/parakeet-tdt_ctc-1.1b | handy-computer | 1.87 | low | |
| 663 | parakeet-ctc-1.1b-gguf | nvidia/parakeet-ctc-1.1b | handy-computer | 1.85 | low | |
| 664 | whisper-large-v3-gguf | openai/whisper-large-v3 | handy-computer | 1.81 | low | |
| 665 | whisper-large-v3-gguf | openai/whisper-large-v3 | memoravox | 1.81 | low | |
| 666 | MOSS-TTS-GGUF | — | OpenMOSS-Team | 1.79 | low | |
| 667 | MOSS-TTS-GGUF | — | niobures | 1.79 | low | |
| 668 | Fun-ASR-Nano-2512-gguf | FunAudioLLM/Fun-ASR-Nano-2512 | handy-computer | 1.78 | low | |
| 669 | Fun-ASR-MLT-Nano-2512-gguf | FunAudioLLM/Fun-ASR-MLT-Nano-2512 | handy-computer | 1.74 | low | |
| 670 | MOSS-TTS | — | OpenMOSS-Team | 1.69 | low | Open‑source |
| 671 | MOSS-TTS-Local-Transformer | OpenMOSS-Team/MOSS-TTS-Local-Transformer | OpenMOSS-Team | 1.69 | low | Open‑source |
| 672 | MOSS-TTS-GGUF | — | OpenMOSS-Team | 1.69 | low | Open‑source |
| 673 | MOSS-TTS-GGUF | — | niobures | 1.69 | low | Open‑source |
| 674 | canary-qwen-2.5b-gguf | nvidia/canary-qwen-2.5b | handy-computer | 1.63 | low | |
| 675 | Qwen3-ASR-1.7B-gguf | Qwen/Qwen3-ASR-1.7B | handy-computer | 1.62 | low | |
| 676 | canary-1b-flash-gguf | nvidia/canary-1b-flash | handy-computer | 1.62 | low | |
| 677 | parakeet-rnnt-0.6b-gguf | nvidia/parakeet-rnnt-0.6b | handy-computer | 1.62 | low | |
| 678 | canary-1b-flash-gguf | nvidia/canary-1b-flash | memoravox | 1.62 | low | |
| 679 | parakeet-rnnt-0.6b-gguf | nvidia/parakeet-rnnt-0.6b | memoravox | 1.62 | low | |
| 680 | spear-xlarge-speech-audio-v2 | — | marcoyang | 1.6 | low | |
| 681 | parakeet-unified-en-0.6b-gguf | nvidia/parakeet-unified-en-0.6b | handy-computer | 1.59 | low | |
| 682 | parakeet-unified-en-0.6b-gguf | nvidia/parakeet-unified-en-0.6b | memoravox | 1.59 | low | |
| 683 | Voxtral-Small-24B-2507-gguf | mistralai/Voxtral-Small-24B-2507 | handy-computer | 1.56 | low | |
| 684 | canary-1b-gguf | nvidia/canary-1b | handy-computer | 1.55 | low | |
| 685 | gepard-1.0 | nineninesix/qwen3_5-full-attn-only-14 | nineninesix | 1.5 | low | |
| 686 | granite-speech-4.1-2b-plus-gguf | ibm-granite/granite-speech-4.1-2b-plus | handy-computer | 1.49 | low | |
| 687 | granite-speech-4.1-2b-plus-gguf | ibm-granite/granite-speech-4.1-2b-plus | memoravox | 1.49 | low | |
| 688 | parakeet-rnnt-1.1b-gguf | nvidia/parakeet-rnnt-1.1b | handy-computer | 1.45 | low | |
| 689 | inflect-v2-coreml | owensong/Inflect-Micro-v2 | FluidInference | 1.43 | low | |
| 690 | granite-4.0-1b-speech-gguf | ibm-granite/granite-4.0-1b-speech | handy-computer | 1.42 | low | |
| 691 | parakeet-tdt-1.1b-gguf | nvidia/parakeet-tdt-1.1b | handy-computer | 1.39 | low | |
| 692 | granite-speech-4.1-2b-gguf | ibm-granite/granite-speech-4.1-2b | handy-computer | 1.31 | low | |
| 693 | granite-speech-4.1-2b-gguf | ibm-granite/granite-speech-4.1-2b | memoravox | 1.31 | low | |
| 694 | asr-whisper-large-v2-salt | openai/whisper-large-v2 | Sunbird | 1.3 | low | |
| 695 | granite-speech-4.1-2b-nar-gguf | ibm-granite/granite-speech-4.1-2b-nar | handy-computer | 1.29 | low | |
| 696 | granite-speech-4.1-2b-nar-gguf | ibm-granite/granite-speech-4.1-2b-nar | memoravox | 1.29 | low | |
| 697 | cohere-transcribe-03-2026-gguf | CohereLabs/cohere-transcribe-03-2026 | handy-computer | 1.26 | low | |
| 698 | cohere-transcribe-03-2026-gguf | CohereLabs/cohere-transcribe-03-2026 | memoravox | 1.26 | low | |
| 699 | wav2vec2-xls-r-1b-japanese-hiragana-katakana | — | AndrewMcDowell | 1.01 | medium | |
| 700 | wav2vec2-vietnamese-asr-v1 | nguyenvulebinh/wav2vec2-base-vietnamese-250h | cheesethuong | 1.0 | medium | |
| 701 | MERaLiON-OmniVoice-Hokkien-TTS | — | MERaLiON | 1.0 | medium | |
| 702 | fastconformer-quran-ar | — | mohammed | 0.38 | low | |