clarification and added new idea

README.md

@@ -15,7 +15,8 @@ Ask YouTube GPT allows you to ask questions about a set of YouTube Videos using
 
 ## Ideas / Future Improvements
 
-- [ ] Add "Suggest a question" feature (maybe through clustering?)
+- [ ] Omit the need for a set of videos, and instead use a search query to find videos on the fly.
+- [ ] Add "Suggest a question" feature given the videos (maybe through clustering?)
 - [ ] Add explainable segment retrieval (i.e. why did that specific segment get chosen to answer the question?)
 - [ ] Add OpenAI embeddings
 - [ ] Improve the retrieval process allowing for the retrieval to return at least one segment from each video (This would in turn enable comparision of videos, i.e. which video explains topic X in more depth?)

app.py

@@ -243,7 +243,7 @@ title = "Ask YouTube GPT 📺"
 with gr.Blocks() as demo:
 
     gr.Markdown(f'<center><h1>{title}</h1></center>')
-    gr.Markdown(f'Ask YouTube GPT allows you to ask questions about a set of YouTube Videos using Topic Segmentation, Universal Sentence Encoding, and Open AI. The returned response cites the video title, author and timestamp in square brackets where the information is located, adding credibility to the responses and helping you locate incorrect information. If you need one, get your Open AI API key <a href="https://platform.openai.com/account/api-keys">here</a>.</p>')
+    gr.Markdown(f'Ask YouTube GPT allows you to ask questions about a set of YouTube Videos using Topic Segmentation, Universal Sentence Encoding, and Open AI. It does not use the video/s itself, but rather the transcript/s of such video/s. The returned response cites the video title, author and timestamp in square brackets where the information is located, adding credibility to the responses and helping you locate incorrect information. If you need one, get your Open AI API key <a href="https://platform.openai.com/account/api-keys">here</a>.</p>')
 
     with gr.Row():