diff --git a/doc/source/demo.rst b/doc/source/demo.rst
index f0ca761..4bb1513 100644
--- a/doc/source/demo.rst
+++ b/doc/source/demo.rst
@@ -1,2 +1,143 @@
Audio Sample
-==================
\ No newline at end of file
+==================
+
+TTS udio samples
+-------------------
+
+Audio samples generated by a TTS system. Text is first transformed into spectrogram
+by a text-to-spectrogram model, then the spectrogram is converted into raw audio by
+a vocoder.
+
+.. raw:: html
+
+
+
+
+Vocoder audio samples
+--------------------------
+
+Audio samples generated from ground-truth spectrograms with a vocoder.
+
+
diff --git a/doc/source/tutorials.rst b/doc/source/tutorials.rst
index 6ce1723..18f0843 100644
--- a/doc/source/tutorials.rst
+++ b/doc/source/tutorials.rst
@@ -25,7 +25,7 @@ The code below show how to use a transformer_tts model. After loading the pretra
>>> from parakeet.models import TransformerTTS
>>> from pathlib import Path
>>> import yacs
-
+>>>
>>> # load the pretrained model
>>> frontend = English()
>>> checkpoint_dir = Path("transformer_tts_pretrained")
@@ -61,12 +61,7 @@ Like the example above, after loading the pretrained ConditionalWaveFlow model,
For more details on how to use the model, please refer the documentation.
-.. raw:: html
-
-
+