Any idea on how to deploy it ?

[Steven Du]

To say I have the trained model, and would like to run a testcase, maybe just on the test_index.txt ?

[pannous]

The test index is used for validation while training,
but you can also try at explicity with "caffe test ..." command
if you want to test it with your own voice try the record python script
( you need the coffee python web app as backend, or quickly insert your own classifier)

On Dec 22, 2014, at 10:25 AM, Steven [e-mail] wrote:

To say I have the trained model, and would like to run a testcase, maybe just on the test_index.txt ?

—
Reply to this email directly or view it on GitHub.

[Mingcong Song]

I have got the trained model by running the script. Could you provide an example to recognize the numbers from audio file?

[pannous]

yes, doku README.md updated

[Steven Du]

Thanks ,FYI

Here is another free corpus

LibriSpeech: an ASR corpus based on public domain audio books

http://www.danielpovey.com/files/2015_icassp_librispeech.pdf

By Dan Povey, the core developer of the state of art ASR system:Kaldi

They provide sets of examples in their ./egs folder on different corpus, some of them are free of charge.