Use a configuration-inclusive revision in the download instructions

Publish the existing ESPnet configuration as config.yaml so normal repository downloads include a Hugging Face download-counting query file. Document the configuration filename and preserve release integrity checks.

README.md

@@ -40,6 +40,7 @@ descriptions spanning speech, music, environmental sound, and mixtures of them.
 |---|---|
 | `base.pt` | Native ESPnet weight-only checkpoint |
 | `train_stage2_qwen3_base.yaml` | Reconstructed checkpoint-compatible model configuration |
+| `config.yaml` | Identical copy of `train_stage2_qwen3_base.yaml` under the standard configuration filename |
 
 `base.pt` is a PyTorch mapping with model state under the top-level `module` key. It
 contains 1,382 tensors: 1,381 BF16 tensors from training and one deterministic FP32
@@ -55,7 +56,7 @@ fields should not be treated as an archival record of every training setting.
 
 ```bash
 hf download espnet/bagpiper \
-  --revision d7b91ab261116973fb65cf44ff9590f6f242323b \
+  --revision 91eb1ca645fe912a7f1b07ce6ff418accf3986cd \
   --local-dir bagpiper-base
 
 echo "c9917c1237bd44ea0ea8b41746e5ec5aecae92273538c9b6e4040d26f69a6521  bagpiper-base/base.pt" \
@@ -80,7 +81,7 @@ sequence interface.
 ## ESPnet runtime
 
 The native consumer is `espnet2/speechlm/bin/inference.py`. Reconstruct the model with
-the supplied training YAML, pair it with the task-appropriate inference YAML from the
+`config.yaml`, pair it with the task-appropriate inference YAML from the
 [Bagpiper recipe](https://github.com/jctian98/espnet/tree/1992821bfb6d68c6636d1b75340b8b2b591bc015/egs2/bagpiper/speechlm1),
 and pass `base.pt` through `--model-checkpoint`.