Qwen3-ASR-1.7B-NL

Home/Models/yuriyvnv/Qwen3-ASR-1.7B-NL

yuriyvnv/Qwen3-ASR-1.7B-NL

kind: original
quant: bf16 (16-bit)
family: qwen
modality: automatic-speech-recognition
root model: Qwen/Qwen3-ASR-1.7B
provider: yuriyvnv
license: apache-2.0
language: nl
context: —
library: transformers
vRAM (GB): —
size on disk: —
downloads: 254
likes: 1
created: —
last modified: 2026-04-27T18:57:13+00:00 (2026)
huggingface: https://huggingface.co/yuriyvnv/Qwen3-ASR-1.7B-NL
tags: transformerssafetensorsqwen3_asrautomatic-speech-recognitionspeechqwen3-asrqwendutchfine-tunedcommon-voicenldataset:fsicoli/common_voice_22_0dataset:fixie-ai/common_voice_17_0dataset:yuriyvnv/synthetic_transcript_ptarxiv:2601.21337base_model:Qwen/Qwen3-ASR-1.7Bbase_model:finetune:Qwen/Qwen3-ASR-1.7Blicense:apache-2.0model-indexendpoints_compatibleregion:us

benchmark	category	score	context	conf	evidence
WER Common Voice	speech-asr	22.0		medium	`Common Voice 22`
WER	speech-asr	6.7	Zero-shot WER \| CER (zero-shot → fine-tuned)	low	`\| Common Voice 17 (test) \| 11,266 \| 6.70 \| 5.37 \| -1.33 (-19.9%) \| 2.05 → 1.69 \|`
WER Common Voice	speech-asr	6.7	Common Voice 17 (test)	medium	`\| Common Voice 17 (test) \| 11,266 \| 6.70 \| 5.37 \| -1.33 (-19.9%) \| 2.05 → 1.69 \|`
WER Common Voice	speech-asr	6.68	Common Voice 22 (test)	medium	`\| Common Voice 22 (test) \| 12,033 \| 6.68 \| 5.28 \| -1.40 (-21.0%) \| 2.04 → 1.68 \|`

none scraped.