Brandon Thomas 33485670dd Remove Python models (see storyteller-ml repo), rework Dockerfile
- TTS and W2L were moved to storyteller-ml. TTS has been hosted there
  for a while, but W2L will need to be reworked.

- Restructure Dockerfile to remove old Python deps.

- Also begin using "cargo-chef". Unsure if this will work to speed things up.
2022-11-14 00:39:41 -05:00
2022-10-09 02:58:16 -04:00
2022-11-07 12:47:38 -05:00
2022-07-28 05:22:48 -04:00
2022-11-13 17:37:12 -05:00
2022-11-13 16:14:30 -05:00
2022-06-26 16:56:19 -04:00
2022-11-07 12:47:38 -05:00
2022-11-07 12:47:38 -05:00
2022-08-04 04:40:22 -04:00
2022-11-13 23:09:31 -05:00

storyteller-rust

Our Rust monorepo, which services our API surface area.

Application overview

  • FakeYou.com is powered by:

    • HTTP APIs:
      • storyteller-web - HTTP + database monolith for FakeYou
    • Asynchronous Jobs:
      • tts-download-job - Async download of models
      • tts-inference-job - Async TTS inference
      • w2l-download-job - Async download of videos and images
      • w2l-inference-job - Async W2L inference
  • StorytellerStream is powered by:

    • HTTP APIs:
      • storyteller-web - HTTP + database monolith for FakeYou
    • Websocket events:
      • websocket-gateway - Websocket to pull down Twitch events.
    • Twitch API integration:
      • twitch-pubsub-subscriber - Dynamically subscribes to Twitch PubSub for "active" users

Schema and API notes

Local development

Local development setup README

Production and Deployment

Production builds and deployment README

Development notes

Development notes README

Dump a MySQL report

mysql -u storyteller \
  -pPasswordHere \
  -h IP_HERE \
  -P 3306 \
  -D storyteller \
  -B \
  -e "select * from voice_clone_requests;" | sed "s/'/\'/;s/\t/\",\"/g;s/^/\"/;s/$/\"/;s/\n//g" > voice_clone_report.csv

TODO

  • Revise this README to be more useful.

Notes / TODOs:

Languages
Rust 59.4%
TypeScript 39.3%
JavaScript 0.7%
Python 0.3%
CSS 0.2%
Other 0.1%