Merge pull request #1627 from vespa-engine/kkraune/text-video

Kkraune/text video

README.md

@@ -205,6 +205,12 @@ exported in [ONNX](https://onnx.ai/)-format for retrieval.
 It features separation of compute from storage and query-time vector similarity de-duping.
 It uses PCA to reduce from 768 to 128 dimensions.
 
+[![logo](/assets/vespa-logomark-tiny.png) Text-video search](text-video-search/) is a notebook that
+downloads a set of videos, converts from `.avi` to `.mp4`, creates CLIP embeddings,
+feeds to Vespa and lets you query the videos in text in a Streamlit application.
+It is a good start for creating a video search application using Vespa!
+
+
 
 ## Ranking
 [![logo](/assets/vespa-logomark-tiny.png) MS Marco Passage Ranking](msmarco-ranking/)

text-video-search/README.md

@@ -11,10 +11,11 @@ Build a text-video search from scratch based on CLIP models with Vespa python AP
 
 [See Animation](https://data.vespa-cloud.com/sample-apps-data/video_demo.gif)
 
-## Create the application from scratch in a Jupyter Notebook
+## Create the application in a Jupyter Notebook
 
 Create, deploy, feed and query the application from [a single notebook](src/python/create-feed-query-text-video-search.ipynb)
-with [the Vespa python API](https://pyvespa.readthedocs.io/en/latest/index.html).
+with [Pyvespa](https://pyvespa.readthedocs.io/en/latest/index.html).
+
 
 ### Demo the search app