Removed LLM compute amortization comment

README.md

@@ -75,9 +75,9 @@ Miso TTS 8B uses two transformer components:
 - A smaller decoder transformer that autoregressively predicts higher-order
   audio codebooks within each frame.
 
-The model follows Sesame's compute-amortized decoder design: codebook 0 is
+Codebook 0 is
 predicted from the backbone hidden state, while codebooks 1 through 31 are
-predicted by the audio decoder.
+predicted by the audio decoder autoregressively in codebook depth.
 
 ---