Skip to content

One-file learning #28

@heinsaar

Description

@heinsaar
  1. Augment a single file into a dataset.
  2. Train an ASR model on the resulting dataset.
  3. Test the ASR model on another dataset.

Key question: what's the best accuracy achievable from a single file augmentation in this way?

Metadata

Metadata

Assignees

No one assigned

    Labels

    bigNew feature or requestexperimentsVarious ideas worth experimenting withmMENTEXRelated to the module or process named in the labelnextImportant next yardstone

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions