Qwen3-ASR-1.7B-hf

Home/Models/Qwen/Qwen3-ASR-1.7B-hf

Qwen/Qwen3-ASR-1.7B-hf

kind: original
quant: bf16 (16-bit)
family: qwen
modality: automatic-speech-recognition
root model: Qwen/Qwen3-ASR-1.7B-hf
provider: Qwen
license: apache-2.0
language: zh, en, yue, ar, de, fr, es, pt, id, it, ko, ru
context: —
library: transformers
vRAM (GB): —
size on disk: —
downloads: 83,563
likes: 43
created: —
last modified: 2026-07-22T11:58:31+00:00 (2026)
huggingface: https://huggingface.co/Qwen/Qwen3-ASR-1.7B-hf
tags: transformerssafetensorsqwen3_asrautomatic-speech-recognitionzhenyueardefresptiditkoruthvijatrhimsnlsv

benchmark	category	score	conf	evidence
WER Earnings-22	speech-asr	9.88	high	`\| Qwen3-ASR-1.7B-hf \| 5.59 \| 9.26 \| 9.88 \| 7.25 \| 1.24 \| 2.92 \| 2.58 \| 5.99 \|`
WER AMI	speech-asr	9.26	high	`\| Qwen3-ASR-1.7B-hf \| 5.59 \| 9.26 \| 9.88 \| 7.25 \| 1.24 \| 2.92 \| 2.58 \| 5.99 \|`
WER GigaSpeech	speech-asr	7.25	high	`\| Qwen3-ASR-1.7B-hf \| 5.59 \| 9.26 \| 9.88 \| 7.25 \| 1.24 \| 2.92 \| 2.58 \| 5.99 \|`
WER VoxPopuli	speech-asr	5.99	high	`\| Qwen3-ASR-1.7B-hf \| 5.59 \| 9.26 \| 9.88 \| 7.25 \| 1.24 \| 2.92 \| 2.58 \| 5.99 \|`
WER	speech-asr	5.59	high	`\| Qwen3-ASR-1.7B-hf \| 5.59 \| 9.26 \| 9.88 \| 7.25 \| 1.24 \| 2.92 \| 2.58 \| 5.99 \|`
WER SPGISpeech	speech-asr	2.58	high	`\| Qwen3-ASR-1.7B-hf \| 5.59 \| 9.26 \| 9.88 \| 7.25 \| 1.24 \| 2.92 \| 2.58 \| 5.99 \|`

none scraped.