HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

whisper-large-v3-turbo-russian-codeswitch

coriollon/whisper-large-v3-turbo-russian-codeswitch

kind
original
quant
bf16 (16-bit)
family
misc
modality
automatic-speech-recognition
root model
coriollon/whisper-large-v3-turbo-russian
provider
coriollon
license
apache-2.0
language
ru, en
context
—
library
transformers
vRAM (GB)
—
size on disk
—
downloads
265
likes
1
created
—
last modified
2026-05-15T07:35:31+00:00 (2026)
huggingface
https://huggingface.co/coriollon/whisper-large-v3-turbo-russian-codeswitch
tags
transformerssafetensorswhisperautomatic-speech-recognitionasrrussiancode-switchingspeech-recognitionfaster-whisperruenbase_model:coriollon/whisper-large-v3-turbo-russianbase_model:finetune:coriollon/whisper-large-v3-turbo-russianlicense:apache-2.0endpoints_compatibleregion:us

Benchmark scores (4)

benchmarkcategoryscorecontextconfevidence
WERspeech-asr24.64WERmedium| WER | 24.64% | **9.08%** | **−15.56 pp** |
WER Common Voicespeech-asr21.0mediumCommon Voice 21
WER LibriSpeechspeech-asr8.5RuLibriSpeechmedium| RuLibriSpeech | 8.50% | 8.57% | +0.07 |
WER Common Voicespeech-asr5.33Common Voice 21 RUmedium| Common Voice 21 RU | 5.33% | 6.87% | +1.54 |

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.