diff --git a/site/en/gemma/docs/core/huggingface_inference.ipynb b/site/en/gemma/docs/core/huggingface_inference.ipynb
index c6b61e05b..426cc66af 100644
--- a/site/en/gemma/docs/core/huggingface_inference.ipynb
+++ b/site/en/gemma/docs/core/huggingface_inference.ipynb
@@ -502,7 +502,7 @@
         "id": "vpp8tZDFU-UJ"
       },
       "source": [
-        "You can include multiple images in your prompt by including additional `\"type\": \"audio\",` entries in the `content` list. If you are prompting with audio data, but without a template, use the `<audio_soft_token>` syntax in the text of your prompt.\n",
+        "You can include multiple audio files in your prompt by including additional `\"type\": \"audio\",` entries in the `content` list. If you are prompting with audio data, but without a template, use the `<audio_soft_token>` syntax in the text of your prompt.\n",
         "\n",
         "Note: Do not use `<audio_soft_token>` tokens in text portion of a prompt template as this approach creates redundant tokens and processing errors. The Transformers library uses `<audio_soft_token>` as a placeholder for images in prompts, rather than `<start_of_audio>` for consistency across models."
       ]