Hello, I'm playing around with speech codecs right now and I'm looking into training codebooks or to be more specific, the 1200bit codec's codebook. >From the code I understand that it is using the split codebooks generated using vq_train_jvm.c file.
What I'm not sure about is the input of the training, I assume they should be a textfile already comprised of lsp's of the training speech data?
_______________________________________________ Freetel-codec2 mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freetel-codec2
