HF Model & Benchmark Matrix

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

Nemotron-Speech-Streaming-0.6B-CoreML-INT8

aufklarer/Nemotron-Speech-Streaming-0.6B-CoreML-INT8

kind
original
quant
bf16 (16-bit)
family
nemotron
modality
automatic-speech-recognition
root model
nvidia/nemotron-speech-streaming-en-0.6b
provider
aufklarer
license
other
language
en
context
—
library
coremltools
vRAM (GB)
—
size on disk
—
downloads
147
likes
0
created
—
last modified
2026-04-22T17:35:28+00:00 (2026)
huggingface
https://huggingface.co/aufklarer/Nemotron-Speech-Streaming-0.6B-CoreML-INT8
tags
coremltoolscoremlspeechasrstreamingapple-siliconneural-enginepunctuationrnn-tfastconformerautomatic-speech-recognitionenbase_model:nvidia/nemotron-speech-streaming-en-0.6bbase_model:finetune:nvidia/nemotron-speech-streaming-en-0.6blicense:otherregion:us

Benchmark scores (9)

benchmarkcategoryscorecontextconfevidence
WER Earnings-22speech-asr12.52Earnings22medium| Earnings22 | 12.52 |
WERspeech-asr11.73high| AMI | 11.73 |
WER AMIspeech-asr11.73AMImedium| AMI | 11.73 |
WER GigaSpeechspeech-asr9.66Gigaspeechmedium| Gigaspeech | 9.66 |
WER VoxPopulispeech-asr7.91VoxPopulimedium| VoxPopuli | 7.91 |
WER LibriSpeechspeech-asr4.84LibriSpeech test-othermedium| LibriSpeech test-other | 4.84 |
WER TED-LIUMspeech-asr3.5TEDLIUMmedium| TEDLIUM | 3.50 |
WER SPGISpeechspeech-asr2.97SPGI Speechmedium| SPGI Speech | 2.97 |
WER LibriSpeechspeech-asr2.32LibriSpeech test-cleanmedium| LibriSpeech test-clean | 2.32 |

Discussions (0)

none scraped.

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