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

Merge pull request #1261 from lissyx/doc-decoder

Browse Source 
Fix doc with no --config=monolithic for libctc_decoder_with_kenlm.so
X-DeepSpeech: NOBUILD
This commit is contained in:
lissyx 2018-02-20 11:32:05 +01:00 committed by GitHub
commit 93393ddca3
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 5 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
native_client/README.md
View File

@ -69,10 +69,11 @@ ln -s ../DeepSpeech/native_client ./
Before building the DeepSpeech client libraries, you will need to prepare your environment to configure and build TensorFlow. Follow the [instructions](https://www.tensorflow.org/install/install_sources) on the TensorFlow site for your platform, up to the end of 'Configure the installation'.
Then you can build the Tensorflow and DeepSpeech libraries.
Then you can build the Tensorflow and DeepSpeech libraries. Please note that the flags for `libctc_decoder_with_kenlm.so` differs a little bit.
```
bazel build --config=monolithic -c opt --copt=-O3 --copt=-fvisibility=hidden //native_client:libdeepspeech.so //native_client:deepspeech_utils //native_client:libctc_decoder_with_kenlm.so //native_client:generate_trie
bazel build -c opt --copt=-O3 //native_client:libctc_decoder_with_kenlm.so
bazel build --config=monolithic -c opt --copt=-O3 --copt=-fvisibility=hidden //native_client:libdeepspeech.so //native_client:deepspeech_utils //native_client:generate_trie
```
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.
@ -97,10 +98,10 @@ Bazel defines:
Bazel targets:
* `//native_client:deepspeech_model`: to produce `libdeepspeech_model.so`
In the end, the previous example becomes:
In the end, the previous example becomes (no change for `libctc_decoder_with_kenlm.so`):
```
bazel build --config=monolithic -c opt --copt=-O3 --copt=-fvisibility=hidden --define=DS_NATIVE_MODEL=1 --define=DS_MODEL_TIMESTEPS=64 --define=DS_MODEL_FRAMESIZE=494 --define=DS_MODEL_FILE=/tmp/model.ldc93s1.pb //native_client:deepspeech_model //native_client:libdeepspeech.so //native_client:deepspeech_utils //native_client:libctc_decoder_with_kenlm.so //native_client:generate_trie
bazel build --config=monolithic -c opt --copt=-O3 --copt=-fvisibility=hidden --define=DS_NATIVE_MODEL=1 --define=DS_MODEL_TIMESTEPS=64 --define=DS_MODEL_FRAMESIZE=494 --define=DS_MODEL_FILE=/tmp/model.ldc93s1.pb //native_client:deepspeech_model //native_client:libdeepspeech.so //native_client:deepspeech_utils //native_client:generate_trie
```
Later, when building either `deepspeech` binaries or bindings, you will have to add some extra variables to your `make` command-line (assuming `TFDIR` points to your TensorFlow's git clone):