-
It's possible to significantly reduce the latency of ASR for systems that support it by doing incremental/"online" recognition: processing audio as it comes in. Especially for interactive use (mobile…
-
I couldn't figure out exactly who to send this to as a lot of people interact with the TEDLIUM data set, so I'm just putting it as an issue on github.
I'm working on some scripts to resegment data, a…
-
A Kaldi STT connection to the server can be closed by the client or the server.
The current implementation handles the case in which the client closes the
connection correctly, but not the case in w…
-
Running `sudo docker pull claritylab/lucida:latest` yields
`failed to register layer: ApplyLayer exit status 1 stdout: stderr: write /usr/local/lucida/lucida/speechrecognition/common/src/gmmbin/gmm-…
-
g++ -msse -msse2 -Wall -I.. -pthread -DKALDI_DOUBLEPRECISION=0 -DHAVE_POSIX_MEMALIGN -Wno-sign-compare -Wno-unused-local-typedefs -Winit-self -DHAVE_EXECINFO_H=1 -rdynamic -DHAVE_CXXABI_H -DHAVE_ATLAS…
-
As of a37daa626a4dc78f7dc9ea8a745aae17a8ae1a15
**Issues**:
- All new services should be integrated into the build system and should pass continuous integration
- OpenEphyra doesn't build, needs a Make…
-
Hi Povey,
I run the same training on the K20 and K80, the CUDA version and nvidia driver version are the same. But I found that the K80 is even slower than the K20 .each iteration for K20 is a…
-
Err http://security.ubuntu.com trusty-security InRelease
Err http://extras.ubuntu.com trusty InRelease
Err http://ppa.launchpad.net trusty InRelease
Err http://ppa.launchpad.net trusty InRelease
E…
-
Hi,
I have Fedora 23 (so I cannot run the Ubuntu VM).
Could you please list the basic install steps to get the the full transcriber setup (The package dependencies I will figure out once I start ins…
-
Hi all, Thanks.
I am builded Kaldi in Ubuntu machine. I searched a lot and there are lot of library and binaries are builded in the directory. I ran the sample program from "egs/voxforge/online_de…