HF Model & Benchmark Matrix

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

parakeet-rnnt-0.6b

nvidia/parakeet-rnnt-0.6b

kind
original
quant
bf16 (16-bit)
family
misc
modality
automatic-speech-recognition
root model
nvidia/parakeet-rnnt-0.6b
provider
nvidia
license
cc-by-4.0
language
en
context
—
library
nemo
vRAM (GB)
—
size on disk
—
downloads
38,179
likes
15
created
—
last modified
2026-07-06T13:52:49+00:00 (2026)
huggingface
https://huggingface.co/nvidia/parakeet-rnnt-0.6b
tags
nemosafetensorsparakeet_rnntfeature-extractionautomatic-speech-recognitionspeechaudioTransducerFastConformerConformerpytorchNeMohf-asr-leaderboardtransformersendataset:librispeech_asrdataset:fisher_corpusdataset:Switchboard-1dataset:WSJ-0dataset:WSJ-1dataset:National-Singapore-Corpus-Part-1dataset:National-Singapore-Corpus-Part-6dataset:vctkdataset:voxpopuli

Benchmark scores (7)

benchmarkcategoryscorecontextconfevidence
WER AMIspeech-asr17.55low| 1.22.0 | SentencePiece Unigram | 1024 | 17.55 | 14.78 | 10.07 | 1.63 | 3.06 | 3.47 | 3.86 | 6.05 | 8.07 |
WER Earnings-22speech-asr14.78low| 1.22.0 | SentencePiece Unigram | 1024 | 17.55 | 14.78 | 10.07 | 1.63 | 3.06 | 3.47 | 3.86 | 6.05 | 8.07 |
WER GigaSpeechspeech-asr10.07low| 1.22.0 | SentencePiece Unigram | 1024 | 17.55 | 14.78 | 10.07 | 1.63 | 3.06 | 3.47 | 3.86 | 6.05 | 8.07 |
WER Common Voicespeech-asr6.05low| 1.22.0 | SentencePiece Unigram | 1024 | 17.55 | 14.78 | 10.07 | 1.63 | 3.06 | 3.47 | 3.86 | 6.05 | 8.07 |
WER VoxPopulispeech-asr3.86low| 1.22.0 | SentencePiece Unigram | 1024 | 17.55 | 14.78 | 10.07 | 1.63 | 3.06 | 3.47 | 3.86 | 6.05 | 8.07 |
WER TED-LIUMspeech-asr3.47low| 1.22.0 | SentencePiece Unigram | 1024 | 17.55 | 14.78 | 10.07 | 1.63 | 3.06 | 3.47 | 3.86 | 6.05 | 8.07 |
WER SPGISpeechspeech-asr3.06low| 1.22.0 | SentencePiece Unigram | 1024 | 17.55 | 14.78 | 10.07 | 1.63 | 3.06 | 3.47 | 3.86 | 6.05 | 8.07 |

Discussions (0)

none scraped.

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