Add model architecture section and diagram

.gitattributes

@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 images/reazonspeech_test.png filter=lfs diff=lfs merge=lfs -text
 images/common_voice_25_ja_test.png filter=lfs diff=lfs merge=lfs -text
 images/fleurs_ja_test.png filter=lfs diff=lfs merge=lfs -text
+images/lambda_audio_architecture.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -55,6 +55,12 @@ On Japanese subset of Google FLEURS ASR benchmark, it achieves a character error
 | Output | Japanese transcription text |
 | Audio feature interval | 20 ms after the encoder; 40 ms after the adapter |
 
+## Architecture
+
+The Zipformer encoder extracts features from 16 kHz audio. An audio adapter reduces the sequence length and projects these features into the decoder's hidden dimension. Gated cross-attention in the LFM2 full-attention blocks lets the decoder use audio features to predict the next text token.
+
+![LambdaAudio architecture: Zipformer encoder, audio adapter, and LFM2 decoder with gated cross-attention](images/lambda_audio_architecture.png)
+
 ## Training data
 
 | Stage | Dataset | Sources and labels |