Transcription of alphanumeric words

Hello,

I’m currently working on my custom ASR system. I would like to enhance it by the transcription of alphanumeric words such as emails, tax identification, dates, ID card number, vehicle number and so on. I did some research on it but until now I did not find any answer.
Does anybody know how to approach this problem or maybe is there someone who is also interested in working on this challenging issue?

Thanks in advance!

I know we got some feedback here of people achieveing something similar, have you searched? I guess it basically come down to fine-tune your dataset with such data.

thank you @lissyx

My plan is to try with fine-tuning (or transfer learning if I would need to append new characters) and adjust the langauage model also.