kaldi-gstreamer-server
kaldi-gstreamer-server copied to clipboard
alumae
Real-time full-duplex speech recognition server, based on the Kaldi toolkit and the GStreamer framwork.
I read source code: https://github.com/alumae/kaldi-gstreamer-server/blob/master/kaldigstserver/worker.py Does it mean if PyKaldi advance_decoding() is invoked but no extra num of frames decoded for long time (> 5s), server will return status=1? After...
Service Temporarily Unavailable occurs Any other English demo link ? Thanks in advance
Dear all, Can someone guide me upto which kaldi version, models are supported with gstreamer server? Kaldi is now compiled with MKL libraries. Models built on this latest version give...
I have run this command, and it is running on localhost with port 8088. python kaldigstserver/master_server.py --port=8088 But to access kaldi server I need to run it with public IP...
``` Exception in thread Thread-28: Traceback (most recent call last): File "/usr/lib/python2.7/threading.py", line 801, in __bootstrap_inner self.run() File "/usr/lib/python2.7/threading.py", line 754, in run self.__target(*self.__args, **self.__kwargs) File "/e/ashwinraju/kaldi_asr/asr/py-kaldi/tests/client.py", line 58, in...
Hi, I want to use nnet3 model instead of nnet2. So, How should I modify the configuration from "librispeech_english_nnet2.yaml" file? Could you provide simple settings? I will appreciate any help,...
Apparently, when the model is trained using audio data with a sample rate other than 16kHz, the decoder fails at decoding audio **at any sample rate**, even when tweaking the...
Hi alumae, I plan to use rnnlm for rescoring the n-best sentences. However, I found that the acoustic scores are not generated and thus a combination of lm score and...
Hi guys, For now I'm trying to get current data inside the pipeline. In the decoding, it takes data chunk sent from worker and in function `process_data` it will push...
First of all, I added cutter into decoder2.py, where I followed how the decoder.py use the cutter. And if I change sample rate in caps_str, for example, from 44.1k to...
