HF Model & Benchmark Matrix

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

parakeet_realtime_eou_120m-v1

nvidia/parakeet_realtime_eou_120m-v1

kind
original
quant
bf16 (16-bit)
family
misc
modality
—
root model
nvidia/parakeet_realtime_eou_120m-v1
provider
nvidia
license
other
language
en
context
—
library
nemo
vRAM (GB)
—
size on disk
—
downloads
486
likes
151
created
—
last modified
2025-12-03T22:28:49+00:00 (2025)
huggingface
https://huggingface.co/nvidia/parakeet_realtime_eou_120m-v1
tags
nemospeech-recognitionFastConformerend-of-utterancevoice agentenarxiv:2305.05084arxiv:2312.17279license:otherregion:us

Benchmark scores (6)

benchmarkcategoryscorecontextconfevidence
WER Earnings-22speech-asr15.76low| WER (%) | 9.30 | 15.62 | 15.76 | 13.31 | 3.61 | 7.79 | 3.79 | 5.48 | 9.07 |
WER AMIspeech-asr15.62low| WER (%) | 9.30 | 15.62 | 15.76 | 13.31 | 3.61 | 7.79 | 3.79 | 5.48 | 9.07 |
WER GigaSpeechspeech-asr13.31low| WER (%) | 9.30 | 15.62 | 15.76 | 13.31 | 3.61 | 7.79 | 3.79 | 5.48 | 9.07 |
WERspeech-asr9.3WER (%)medium| WER (%) | 9.30 | 15.62 | 15.76 | 13.31 | 3.61 | 7.79 | 3.79 | 5.48 | 9.07 |
WER VoxPopulispeech-asr9.07low| WER (%) | 9.30 | 15.62 | 15.76 | 13.31 | 3.61 | 7.79 | 3.79 | 5.48 | 9.07 |
WER TED-LIUMspeech-asr5.48low| WER (%) | 9.30 | 15.62 | 15.76 | 13.31 | 3.61 | 7.79 | 3.79 | 5.48 | 9.07 |

Discussions (0)

none scraped.

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