for (int i = 0; i <16000; i++){
    for (int j = 0; j <121; j++){
    {
    if (j > i){
    data_out[out] += filter[j] * data_in[j - i];
    }
  }
}

memcpy(&filter_state_save[0], data_in[16000-120], 120*sizeof(data_in));