Value of silence_weight

Hi Nickolay!
I've started using the vosk-api, and I've had a few utterances where the output was worse compared to when doing decoding with kaldi's tcp binary (stuff like no instead of yes being recognized). I managed to track down what was causing the difference for most (unfortunately not all) to the silence_weight, which is [hardcoded to 1e-3](https://github.com/alphacep/vosk-api/blob/ce5ffb980a2cc7f0d631e05849efac61a0c45b50/src/model.cc#L230) in vosk.

As I understand training is done without this silence weight being used ([source](https://groups.google.com/g/kaldi-help/c/A73Gw2hhL30)), so wouldn't it make more sense to set that to 1.? What do you think?