whisper-large-v3-turbo-lt

Home/Models/svogunas/whisper-large-v3-turbo-lt

svogunas/whisper-large-v3-turbo-lt

kind: original
quant: bf16 (16-bit)
family: misc
modality: automatic-speech-recognition
root model: openai/whisper-large-v3-turbo
provider: svogunas
license: cc-by-4.0
language: lt
context: —
library: transformers
vRAM (GB): —
size on disk: —
downloads: 279
likes: 1
created: —
last modified: 2026-06-22T12:04:40+00:00 (2026)
huggingface: https://huggingface.co/svogunas/whisper-large-v3-turbo-lt
tags: transformerssafetensorswhisperautomatic-speech-recognitionlithuanianlietuviuasrspeech-recognitionliepaltbase_model:openai/whisper-large-v3-turbobase_model:finetune:openai/whisper-large-v3-turbolicense:cc-by-4.0model-indexendpoints_compatibleregion:us

benchmark	category	score	context	conf	evidence
WER	speech-asr	49.66		low	\| `openai/whisper-large-v3-turbo` (base) \| 49.66 % \| — \|
WER Common Voice	speech-asr	32.93	Common Voice 17 lt (test, n=4 753)	medium	`\| Common Voice 17 lt (test, n=4 753) \| 32.93 % \| 9.49 % \| −71 % \|`
WER FLEURS	speech-asr	25.5	FLEURS-lt (test, n=986)	medium	`\| FLEURS-lt (test, n=986) \| 25.50 % \| 13.18 % \| −48 % \|`
WER Common Voice	speech-asr	17.0		medium	`Common Voice 17`
CER	speech-asr	1.56		low	`\| this fine-tune \| 6.59 % \| 1.56 % \|`

none scraped.