Offline Adaption

Home Forums OpenEars Offline Adaption

Viewing 2 posts - 1 through 2 (of 2 total)

  • Author
    Posts
  • #1032867

    wjm4
    Participant

    I tried to add this to the adaptation topic but it is closed.

    Is there a way to do adaption away from the device and then transfer it to the device?

    If not, what is the best way to improve recognition for a small set of vocabulary (numbers 1-100)?

    #1032868

    Halle Winkler
    Politepix

    Hi,

    Adaptation advice is unfortunately outside of the support offered for OpenEars since it isn’t an OpenEars feature, and advice you get from the CMU Sphinx project may not necessarily be applicable to the specific OpenEars acoustic models you want to use. However, if this potential outcome is OK for you and/or you are knowledgable enough or up for enough self-guided research to get to grips with this advanced topic in ASR, I recommend bringing this question over to the CMU Sphinx forums where they maintain adaptation-related tooling.

    I don’t have advice about optimizing number recognition beyond the topics where it has already been discussed here – it is a known issue for offline recognition. I recommend a search for “number” in the forum topic search field to read past discussions. Good luck!

    • This reply was modified 3 weeks, 3 days ago by  Halle Winkler.
Viewing 2 posts - 1 through 2 (of 2 total)
  • The topic ‘Offline Adaption’ is closed to new replies.