Custom Speech Model Training RRS feed

  • Question

  • Hello,

    At the moment we can train a new model using a data source - related text file or audio + text files, but one one file, either a zip or txt.

    In an enterprise scenario, new data will keep getting generated. So to add this to the training do we have to create a new file with all the data and train a new model everytime or is there some way to train an existing model with the additional data?



    Tuesday, June 25, 2019 12:35 PM

All replies