Collected sources and patterns will appear here. Add from search, explore, or the patterns library.
CTCPosteriors, WFSTGraph -> Transcription
Decode neural network CTC posteriors over a unified Token-Lexicon-Grammar search graph to incorporate custom language models.
Problem it solves
Pure end-to-end models cannot easily incorporate rare domain vocabulary or external rule-based language constraints.
Consumes
Emits
The real projects this mechanism was found in. Attribution is the point — this is how the best teams actually do it.