Kaldi e2e model support?

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

Kaldi e2e model support? #221

Open Umar17 opened 4 years ago

Umar17 commented 4 years ago

Hello,

Is kaldi gstreamer support kaldi e2e model (which doesn't use i-vectors for training and thus no ivector extractor is trained). Additionally for acoustic model, kaldi gstreamer raises error for e2e model: Expected token "", got instead "".

any help please.

Best Regards

alumae commented 4 years ago

The e2e model should be usable as other Kaldi chain models. Just don't specify the i-vetor extractor. Make sure that you are linking to recent Kaldi.