search

alumae / kaldi-gstreamer-server

Real-time full-duplex speech recognition server, based on the Kaldi toolkit and the GStreamer framwork.
BSD 2-Clause "Simplified" License
1.07k stars 339 forks source link

not printed greek characters #217

Open nikosapaster opened 5 years ago

nikosapaster commented 5 years ago

I made an example with english characters and greek audio on kaldi which was working fine with kaldi -streamer-server. After this test I tried greek characters, models seems to be trained but kaldi-gstreamer-server does not print result. Are there any guidelines to make it work with greek or is this an issue?

naxingyu commented 5 years ago

the server is ok with utf-8 encoded charactors.

nikosapaster commented 5 years ago

I use ubuntu 18.04 LTS, docker-kaldi-gstreamer-server and I send the audio from the host to the docker image. Inform me if there are any configurations needed to support greek because as it is something is wrong.

alumae commented 5 years ago

Does this happen when we use client.py? Make sure that the terminal you use client.py in, is configured for UTF-8. This can be usually accomplished with export LC_ALL=en_US.UTF-8 or something similar.

naxingyu commented 5 years ago

It might due to incorrect processing of the result message in your host. Debug your client.

qusaiobaida commented 5 years ago

Hi, I had the same problem with Arabic characters and fixed it

just go to kaldigstserver/worker.py and add the following: import sys reload(sys) sys.setdefaultencoding('utf-8')

and the client will work fine :)