Menu

question about generic model hub4wsj_sc_8k

Help
tfpeach
2015-06-08
2015-06-08
  • tfpeach

    tfpeach - 2015-06-08

    Hi, dear all,

    I am using hub4wsj_sc_8k generic model to do some speech recognition. But here is a question. Do I need downsample all of my input files to 8k sample rate? (The original is 16k)

    Thank you.

     
    • bic-user

      bic-user - 2015-06-08

      It's not really necessary. There are max and min frequency in feat.params files. mel filters are built and used only for that space, so it should be ok. It is suggested to use one of default en-us models though.

       

Log in to post a comment.