Tutorial / Example - Fine grained search of the Kubeflow youtube channel content

[jbottum]

The Kubeflow Community could use a current end-to-end example. Current examples: https://www.kubeflow.org/docs/started/kubeflow-examples/

We might be able to create an example that the Community members could use to find (short - 20 second) answers from Community discussions. For example, the Community holds Working Group meetings, which are recorded. If these meetings were saved on a Youtube channel, then we could create a fine grained search option that would find and serve answers to questions submitted to in a web form. Some search example might be: "multi-user pipelines", "unified training operator", or "upgrading Kubeflow".

An example of this solution is explained in this post: https://betterprogramming.pub/fixing-youtube-search-with-openais-whisper-90bb569073cf . It uses OpenAI's Whisper for transcoding and S-BERT for Question/Answer, with streamline or gradio for the webapp.

I am looking for interested parties to help develop this example. Related ideas - If we could get the example functional, we might be able to use it at conferences as a Kubeflow tutorial / workshop i.e. AI at KubeCon EMEA in 2023 and/or as an example for the Kubeflow 1.7 release.

@kubeflow/wg-notebooks-leads @kubeflow/wg-pipeline-leads @kubeflow/wg-training-leads @kubeflow/wg-automl-leads @kubeflow/release-managers

/kind feature /priority p2 /area help wanted

[google-oss-prow[bot]]

@jbottum: The label(s) area/help, area/wanted cannot be applied, because the repository doesn't have them.

In response to this:

The Kubeflow Community could use a current end-to-end example. Current examples: https://www.kubeflow.org/docs/started/kubeflow-examples/

We might be able to create an example that the Community members could use to find (short - 20 second) answers from Community discussions. For example, the Community holds Working Group meetings, which are recorded. If these meetings were saved on a Youtube channel, then we could create a fine grained search option that would find and serve answers to questions submitted to in a web form. Some search example might be: "multi-user pipelines", "unified training operator", or "upgrading Kubeflow".

An example of this solution is explained in this post: https://betterprogramming.pub/fixing-youtube-search-with-openais-whisper-90bb569073cf . It uses OpenAI's Whisper for transcoding and S-BERT for Question/Answer, with streamline or gradio for the webapp.

I am looking for interested parties to help develop this example. Related ideas - If we could get the example functional, we might be able to use it at conferences as a Kubeflow tutorial / workshop i.e. AI at KubeCon EMEA in 2023 and/or as an example for the Kubeflow 1.7 release.

@kubeflow/wg-notebooks-leads @kubeflow/wg-pipeline-leads @kubeflow/wg-training-leads @kubeflow/wg-automl-leads @kubeflow/release-managers

/kind feature /priority p2 /area help wanted

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[varodrig]

Sounds great! Looking for more comments from the community

cc @kubeflow/kubeflow-steering-committee @kubeflow/wg-training-leads @kubeflow/wg-pipeline-leads @kubeflow/wg-notebooks-leads @kubeflow/wg-manifests-leads @kubeflow/release-team @kubeflow/wg-data-leads @kubeflow/wg-deployment-leads

[varodrig]

/help

[google-oss-prow[bot]]

@varodrig: This request has been marked as needing help from a contributor.

Please ensure the request meets the requirements listed here.

If this request no longer meets these requirements, the label can be removed by commenting with the /remove-help command.

In response to this:

/help

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[varodrig]

/area website

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[github-actions[bot]]

This issue has been automatically closed because it has not had recent activity. Please comment "/reopen" to reopen it.