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 341 forks source link

Retrieving word level time stamp #234

Closed shaheenkdr closed 4 years ago

shaheenkdr commented 4 years ago

Is there a way we could retrieve the world level timestamps along with the output?

alumae commented 4 years ago

Set the property word-boundary-file=dir/word_boundary.int (under the decoder conf), this activates word alignment generation.

shaheenkdr commented 4 years ago

Thanks @alumae , I can see the output in worker.log . However which endpoint passes that output back to the client ? Couldnt see it from Python client as well as curl . Would be helpful if you could share some pointers in that regard : /

alumae commented 4 years ago

It is passed in the JSON representation that the client receives from the server.

shaheenkdr commented 4 years ago

both in the websocket as well as the dynamic recognize http end points ?