diff --git a/README.md b/README.md
index b1ec8fc38..217c8e744 100644
--- a/README.md
+++ b/README.md
@@ -72,6 +72,8 @@ The best WER we currently have is:
 |-----|------------|------------|
 | WER | 3.16       | 7.71       |
 
+We provide a Colab notebook to run a pre-trained RNN-T conformer model: [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1_u6yK9jDkPwG_NLrZMN2XK7Aeq4suMO2?usp=sharing)
+
 
 ### Aishell
 
diff --git a/egs/librispeech/ASR/RESULTS.md b/egs/librispeech/ASR/RESULTS.md
index 8d7c867c0..4e62af2c3 100644
--- a/egs/librispeech/ASR/RESULTS.md
+++ b/egs/librispeech/ASR/RESULTS.md
@@ -1,5 +1,51 @@
 ## Results
 
+### LibriSpeech BPE training results (RNN-T)
+
+#### 2021-12-17
+
+RNN-T + Conformer encoder
+
+The best WER is
+
+|     | test-clean | test-other |
+|-----|------------|------------|
+| WER | 3.16       | 7.71       |
+
+using `--epoch 26 --avg 12` during decoding.
+
+The training command to reproduce the above WER is:
+
+```
+export CUDA_VISIBLE_DEVICES="0,1,2,3"
+
+./transducer/train.py \
+  --world-size 4 \
+  --num-epochs 30 \
+  --start-epoch 0 \
+  --exp-dir transducer/exp-lr-2.5-full \
+  --full-libri 1 \
+  --max-duration 250 \
+  --lr-factor 2.5
+```
+
+The decoding command is:
+
+```
+epoch=26
+avg=12
+
+./transducer/decode.py \
+  --epoch $epoch \
+  --avg $avg \
+  --exp-dir transducer/exp-lr-2.5-full \
+  --bpe-model ./data/lang_bpe_500/bpe.model \
+  --max-duration 100
+```
+
+You can find the tensorboard log at: <https://tensorboard.dev/experiment/PYIbeD6zRJez1ViXaRqqeg/>
+
+
 ### LibriSpeech BPE training results (Conformer-CTC)
 
 #### 2021-11-09
diff --git a/egs/librispeech/ASR/transducer/pretrained.py b/egs/librispeech/ASR/transducer/pretrained.py
index 9dedfc16f..bb59a7338 100755
--- a/egs/librispeech/ASR/transducer/pretrained.py
+++ b/egs/librispeech/ASR/transducer/pretrained.py
@@ -276,6 +276,8 @@ def main():
             hyp = beam_search(
                 model=model, encoder_out=encoder_out_i, beam=params.beam_size
             )
+        else:
+            raise ValueError(f"Unsupported method: {params.method}")
 
         hyps.append(sp.decode(hyp).split())