{"apiVersion":"v1","taxonomyVersion":1,"generatedAt":"2026-10-29T15:04:13.464Z","data":{"recordId":"wsa_44np7x67fixr","score":1,"confidence":"high","capped":false,"capReason":null,"sources":[{"n":1,"url":"https://huggingface.co/TeamUNIVA/qwen3_asr_1.7b_ko","publisher":"Hugging Face","authority":"owner-controlled","verification":"verified","lastRetrievedAt":"2026-10-28T00:15:38.667Z","detectedPublishedAt":"2026-10-12T00:49:13.000Z","rights":"original"}],"claims":[{"id":"4ebf6101-5091-4483-b8cf-be8304862229","kind":"capability","statement":"The model is fine-tuned for Korean automatic speech recognition.","citations":[{"source":1,"supportProbability":0.97,"supports":true,"quote":{"text":"This model is a fine-tune of `Qwen/Qwen3-ASR-1.7B` on Korean speech data for automatic speech recognition (ASR), and the successor to [`TeamUNIVA/qwen3_asr_1.7b_ko_beta`](https://huggingface.co/TeamUNIVA/qwen3_asr_1.7b_ko_beta).","rights":"original"}}]},{"id":"ec830cbe-88e0-4035-8409-6bee8e670ec6","kind":"capability","statement":"It achieves a Korean total CER of 0.0483 on the reported benchmarks.","citations":[{"source":1,"supportProbability":0.93,"supports":true,"quote":null}]},{"id":"3973c852-b939-447c-9d41-4921a08b072b","kind":"capability","statement":"Its English LibriSpeech WER is 0.0277, compared with 0.0250 for the base model.","citations":[{"source":1,"supportProbability":0.97,"supports":true,"quote":{"text":"It achieves lower error rates than both the base model and the beta model across Korean benchmarks, with a small degradation on English (LibriSpeech, +0.27 %p WER vs.","rights":"original"}}]},{"id":"1c465148-b1db-43f9-b7cc-b86674a52712","kind":"availability","statement":"The model can be loaded with Qwen3ASRModel.from_pretrained using the TeamUNIVA/qwen3_asr_1.7b_ko identifier.","citations":[{"source":1,"supportProbability":0.96,"supports":true,"quote":{"text":"Starting from `TeamUNIVA/qwen3_asr_1.7b_ko_beta` (itself an SFT-only fine-tune of `Qwen/Qwen3-ASR-1.7B`), the model was trained with full-parameter updates in four sequential stages, using [ms-swift](https://github.com/modelscope/ms-swift):","rights":"original"}}]},{"id":"294d98a0-a93b-4841-9629-93da7f95ced2","kind":"recommendation-relevant","statement":"The model’s performance on other languages may be similar to the base model or require separate validation.","citations":[{"source":1,"supportProbability":0.97,"supports":true,"quote":{"text":"- This model was fine-tuned specifically for Korean speech recognition; performance on other languages may be similar to the base model or require separate validation.","rights":"original"}}]}]},"nextCursor":null,"license":{"catalog":{"name":"CC BY 4.0","spdx":"CC-BY-4.0","url":"https://creativecommons.org/licenses/by/4.0/","attribution":"wowsuchai","attributionUrl":"https://wowsuchai.withglint.app","covers":["catalog metadata","classifications","generated summaries","feed event metadata"]},"excluded":{"covers":["source quotations","linked source content"],"terms":"Not licensed by wowsuchai. Source quotations and linked source content keep their original rights."}}}