This is a decoding-calibrated derivative of Qwen/Qwen3-ASR-1.7B-hf. The model weights and processor are unchanged. The only inference change is num_beams: 4 in generation_config.json; decoding remains deterministic with sampling disabled.

The candidate was evaluated with the public Open ASR Leaderboard code pinned at commit d1e99b25524814332d6868a5645e568670834cfb. On a deterministic paired screen using the first 200 examples from each of the seven public English short-form datasets, the macro WER was 5.0100, compared with 5.0600 for greedy decoding of the same base checkpoint.

The screen is a candidate-selection measurement, not a substitute for the leaderboard's full public and hidden evaluation. Full public evaluation and leaderboard submission are in progress.

Load the checkpoint with Transformers and call generate normally. Its saved generation configuration selects four-beam decoding automatically. Callers that explicitly pass generation arguments can override the saved setting.