Merge branch 'master' of github.com:georgesterpu/avsr-tf1

leave space after phoneme units in the output file

update default flags

upload phone and viseme vocabularies

update launch scripts

allow video as main input for lipreading, controllable through the input_modality flag

Create FUNDING.yml

clean up tcdtimit-related code

update readme

update remaining files from devel repo

update dataset writer

update cells

update files

removed deprecated examples

add files from devel repo