whisper-large-v3-turbo-darija

Home/Models/anaszil/whisper-large-v3-turbo-darija

anaszil/whisper-large-v3-turbo-darija

kind: original
quant: bf16 (16-bit)
family: misc
modality: automatic-speech-recognition
root model: openai/whisper-large-v3-turbo
provider: anaszil
license: mit
language: ar
context: —
library: peft
vRAM (GB): —
size on disk: —
downloads: 1,514
likes: 9
created: —
last modified: 2025-11-09T18:23:18+00:00 (2025)
huggingface: https://huggingface.co/anaszil/whisper-large-v3-turbo-darija
tags: peftsafetensorsautomatic-speech-recognitionarbase_model:openai/whisper-large-v3-turbobase_model:adapter:openai/whisper-large-v3-turbolicense:mitregion:us

benchmark	category	score	context	conf	evidence
WER	speech-asr	24.9		medium	`WER = 24.9 %`
WER	speech-asr	24.88	Word Error Rate (WER)	medium	`\| Word Error Rate (WER) \| 24.88 % \|`
CER	speech-asr	8.3		medium	`CER = 8.3 %`
CER	speech-asr	8.28	Character Error Rate (CER)	medium	`\| Character Error Rate (CER) \| 8.28 % \|`

none scraped.