Diné Bizaad DataBase Vault

Nyzhon Studios LLC

Diné Bizaad DataBase Vault

We are wanting to build a strong foundation for Digitizing and creating tools for digitizing The Navajo Language for Model Training.

Diné College

United States

Hybrid of for-profit and nonprofit

The Gap of rich digitized Data to no data at all being reduced significantly.

We have two solutions we want to work with that include building a model. First is creating a Scan Image/PDF to Digital Text for Diné Bizaad, not this is not a translation tool, it is purely transcription. and sound model we wish to create is a Voice to Digital Text model, same thing, Transcription only. This will revolutionize our way of collecting data for Training model purposes and other uses case our tribe will greatly use.

To this day, there is no proper Scan Image/PDF to Digital Text or Voice to Digital Text tool. These two tools alone will push our Tribe leap forward. This allows for scanning documents that contain or is all Navajo Text to be more seamless, which will encourage more people to speaḱwrite more in navajo. Same with the Voice to Digital Text tool. Sometimes meetings are held in navajo and this tool will help for those wanting to follow along with a transcription tool. Navajo Had a few diacritic marks that other model struggle to transcript á ą ą́ ł a’ ą́ą’