wav2vec2-xlsr-300m-finnish-lm

Home/Models/Finnish-NLP/wav2vec2-xlsr-300m-finnish-lm

Finnish-NLP/wav2vec2-xlsr-300m-finnish-lm

kind: original
quant: bf16 (16-bit)
family: misc
modality: automatic-speech-recognition
root model: —
provider: Finnish-NLP
license: apache-2.0
language: —
context: —
library: —
vRAM (GB): —
size on disk: —
downloads: 3,413
likes: 0
created: —
last modified: 2025-04-08T11:13:43+00:00 (2025)
huggingface: https://huggingface.co/Finnish-NLP/wav2vec2-xlsr-300m-finnish-lm
tags: transformerspytorchsafetensorswav2vec2automatic-speech-recognitionfifinnishgenerated_from_trainerhf-asr-leaderboardrobust-speech-eventdataset:mozilla-foundation/common_voice_7_0arxiv:2111.09296doi:10.57967/hf/2253license:apache-2.0model-indexendpoints_compatibleregion:us

benchmark	category	score	context	conf	evidence
WER	speech-asr	8.16		high	`\| Finnish-NLP/wav2vec2-xlsr-300m-finnish-lm \| 300 million \| 8.16 \| 17.92 \| 1.97 \| 3.36 \|`
WER VoxPopuli	speech-asr	7.97	[VoxPopuli Finnish](https://github.com/facebookresearch/voxpopuli)	medium	`\| [VoxPopuli Finnish](https://github.com/facebookresearch/voxpopuli) \| 21.97 h \| 7.97 % \|`
WER Common Voice	speech-asr	7.0		medium	`Common Voice 7.0`
WER VoxPopuli	speech-asr	5.85	Finnish-NLP/wav2vec2-base-fi-voxpopuli-v2-finetuned	medium	`\| Finnish-NLP/wav2vec2-base-fi-voxpopuli-v2-finetuned \| 95 million \| 5.85 \| 13.52 \| 1.35 \| 2.44 \|`
WER VoxPopuli	speech-asr	4.13	Finnish-NLP/wav2vec2-large-uralic-voxpopuli-v2-finnish	medium	`\| Finnish-NLP/wav2vec2-large-uralic-voxpopuli-v2-finnish \| 300 million \| 4.13 \| 9.66 \| 0.90 \| 1.66 \|`
WER Common Voice	speech-asr	3.52	[Common Voice 7.0 Finnish train + evaluation + other splits](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0)	medium	`\| [Common Voice 7.0 Finnish train + evaluation + other splits](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0) \| 9.70 h \| 3.52 % \|`
CER	speech-asr	1.97		high	`\| Finnish-NLP/wav2vec2-xlsr-300m-finnish-lm \| 300 million \| 8.16 \| 17.92 \| 1.97 \| 3.36 \|`

none scraped.