mirrors/DeepSpeech
2
0
Fork 0
mirror of https://github.com/mozilla/DeepSpeech.git synced 2025-10-26 11:19:39 +00:00
Code Issues Packages Projects Releases Wiki Activity
8d21d0cbd2
DeepSpeech/native_client
History
Alexandre Lissy feebc7d058 Limit libdeepspeech -> tensorflow deps 
On my desktop, makes a `bazel build -c opt [...] //native_client:*`
after a `bazel clean` running in ~150 secs instead of 560 secs.

Fixes #475
2017-03-29 12:32:57 +02:00
..
c_speech_features Implement a native client using TensorFlow C API 2017-02-23 16:35:44 +00:00
kiss_fft130 Implement a native client using TensorFlow C API 2017-02-23 16:35:44 +00:00
BUILD Limit libdeepspeech -> tensorflow deps 2017-03-29 12:32:57 +02:00
client.cc Replace sox_open_memstream_write() for OS-X. Fixes issue #461 2017-03-28 09:26:40 +02:00
deepspeech.cc Pass correct input_lengths value as a Tensor to Session::Run() 2017-03-29 11:39:23 +02:00
deepspeech.h Implement a native client using TensorFlow C API 2017-02-23 16:35:44 +00:00
Makefile Addressed review comment 2017-03-27 17:48:19 +02:00
multistrap.conf Support RPi toolchain cross-compilation of native client 2017-03-22 15:50:54 +01:00
README.md Separate libdeepspeech/c_speech_features/kissfft library builds 2017-03-22 13:30:55 +00:00

README.md

DeepSpeech native client

A native client for running queries on an exported DeepSpeech model.

Requirements

Preparation

Create a symbolic link in the TensorFlow checkout to the deepspeech native_client directory.

cd tensorflow
ln -s ../DeepSpeech/native_client ./

Building

Before building the TensorFlow stand-alone library, you will need to prepare your environment to configure and build TensorFlow. Follow the instructions on the TensorFlow site for your platform, up to the end of 'Configure the installation'.

To build the TensorFlow library, execute the following command:

bazel build -c opt //tensorflow:libtensorflow.so

Then you can build the DeepSpeech native library.

bazel build -c opt //native_client:*

Finally, you can change to the native_client directory and use the Makefile. By default, the Makefile will assume there is a TensorFlow checkout in a directory above the DeepSpeech checkout. If that is not the case, set the environment variable TFDIR to point to the right directory.

cd ../DeepSpeech/native_client
make deepspeech

Running

The client can be run via the Makefile. The client will accept audio of any format your installation of SoX supports.

ARGS="/path/to/output_graph.pb /path/to/audio/file.ogg" make run