HF Model & Benchmark Matrix

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

stt_en_fastconformer_hybrid_large_pc

nvidia/stt_en_fastconformer_hybrid_large_pc

kind
original
quant
bf16 (16-bit)
family
misc
modality
automatic-speech-recognition
root model
nvidia/stt_en_fastconformer_hybrid_large_pc
provider
nvidia
license
cc-by-4.0
language
en
context
—
library
nemo
vRAM (GB)
—
size on disk
—
downloads
2,731
likes
4
created
—
last modified
2025-02-27T13:07:05+00:00 (2025)
huggingface
https://huggingface.co/nvidia/stt_en_fastconformer_hybrid_large_pc
tags
nemoautomatic-speech-recognitionspeechaudioTransducerFastConformerCTCTransformerpytorchNeMohf-asr-leaderboardendataset:librispeech_asrdataset:fisher_corpusdataset:mozilla-foundation/common_voice_11_0dataset:National-Singapore-Corpus-Part-1dataset:vctkdataset:spgidataset:VoxPopulidataset:Europarl-ASRdataset:Multilingual-LibriSpeecharxiv:2305.05084license:cc-by-4.0model-index

Benchmark scores (3)

benchmarkcategoryscorecontextconfevidence
WER MLSspeech-asr4.48low| 1.18.0 | SentencePiece Unigram | 1024 | 7.39 | 8.23 | 4.48 | 4.53 | 4.22 | 4.54 | 9.69 | 8.02 | 10.53 | 10.34 | 2.32 | 2.26 | 1.74 | 2.03 | 4.02 | 4.07 | 4.71 | 4.6 |
WER VoxPopulispeech-asr4.22low| 1.18.0 | SentencePiece Unigram | 1024 | 7.39 | 8.23 | 4.48 | 4.53 | 4.22 | 4.54 | 9.69 | 8.02 | 10.53 | 10.34 | 2.32 | 2.26 | 1.74 | 2.03 | 4.02 | 4.07 | 4.71 | 4.6 |
WER LibriSpeechspeech-asr1.74low| 1.18.0 | SentencePiece Unigram | 1024 | 7.39 | 8.23 | 4.48 | 4.53 | 4.22 | 4.54 | 9.69 | 8.02 | 10.53 | 10.34 | 2.32 | 2.26 | 1.74 | 2.03 | 4.02 | 4.07 | 4.71 | 4.6 |

Discussions (0)

none scraped.

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