Merge pull request #143 from MycroftAI/feature/snap

Add snap packaging recipe

Author: forslund

SNAPCRAFT.md

@@ -0,0 +1,18 @@
+# Mycroft-Precise Snap package
+
+## Package details
+The package is a fairly standard configuration using the Snap [Python Plugin](https://snapcraft.io/docs/python-plugin).
+
+The package version is extracted from the `setup.py` script to remain consistent with the pip information. In addition a comparison with the repo is made to see if this commit matches the latest release commit. If it doesn't match _-dev_ is appended to the version info.
+
+Since Precise uses pyaudio, ALSA is required for the audio to work. This Snap uses the [alsa-mixin](https://snapcraft-alsa.readthedocs.io/en/latest/snapcraft_usage.html) to setup the ALSA components. This results in some minor issues that need to be handled. Library conflicts may occur if audio packages are included as part of the main application part, making them appear twice in conflicting versions. This is the reason that libportaudio2 and pulseaudio are in the stage-packages for the alsa-mixin part.
+
+This package also includes the current pretrained "hey mycroft" model accessed through `/snap/mycroft-precise/current/hey-mycroft/hey-mycroft.pb`.
+
+## Plugs
+Plugs allow the Snap to connect to specific aspects of the host system. Without any specified plugs the application will run entirely within a read-only file system dedicated to the application. It will not be able to access the system, including audio devices, that are outside of the Snap container.
+
+**Plugs used:**
+- home: access to the user's home directory (excluding hidden files)
+- audio-record: Access to the system's audio input devices
+- audio-playback: Access to the system's audio output devices

snap/snapcraft.yaml

@@ -0,0 +1,182 @@
+name: mycroft-precise
+adopt-info: mycroft-precise
+license: Apache-2.0
+summary: A wake word listener from Mycroft AI
+description: >
+  A lightweight, simple-to-use, RNN wake word listener, including all the tools to train your own custom wake word.
+  
+  Mycroft Precise monitors an audio stream (usually a microphone) and when it recognizes a specific phrase it triggers an event. For example, by default users of the Mycroft Voice Assistant are using a Precise model trained to spot the phrase "Hey Mycroft". When Precise recognizes this phrase it puts the Mycroft Voice Assistant into command mode performing speech recognition on whatever is next said by the person using the device.
+  
+  Mycroft Precise is fully open source and can be trained to recognize any short-phrase or sound, from a name to a cough.
+  
+  The default "Hey Mycroft" model is included, to test it, try:
+  `mycroft-precise.listen /snap/mycroft-precise/current/hey-mycroft/hey-mycroft.pb`
+  
+
+  **USAGE**
+  
+  
+  Running the listener
+  
+    mycroft-precise - Alias for mycroft-precise.engine
+    mycroft-precise.engine - Run a model on raw audio data from stdin
+    mycroft-precise.listen - Run a model on microphone audio input
+    mycroft-precise.listen-pocketsphinx - Run the PocketSphinx listener
+  
+  Data collection
+  
+    mycroft-precise.collect - Record audio samples for use with Precise
+    mycroft-precise.add-noise - Create a duplicate dataset with added noise
+  
+  Training
+  
+    mycroft-precise.train - Train a new model on a dataset
+    mycroft-precise.train-generated - Train a model on infinitely generated batches
+    mycroft-precise.train-incremental - Train a model to inhibit activation by marking false activations and retraining
+    mycroft-precise.train-optimize - Use black box optimization to tune model hyperparameters
+    mycroft-precise.train-sampled - Train a model, sampling data points with the highest loss from a larger dataset
+  
+  Evaluation and analysis
+  
+    mycroft-precise.test - Test a model against a dataset
+    mycroft-precise.test-pocketsphinx - Test PocketSphinx against a dataset
+    mycroft-precise.eval - Evaluate a list of models on a dataset
+    mycroft-precise.calc-threshold - Update the threshold values of a model for a dataset to make the sensitivity more accurate and linear
+    mycroft-precise.graph - Show ROC curves for a series of models
+    mycroft-precise.simulate - Simulate listening to long chunks of audio to find unbiased false positive metrics
+  
+  Model conversion
+  
+    mycroft-precise.convert - Convert wake-word model from Keras to TensorFlow
+base: core18
+grade: stable
+confinement: strict
+build-packages: [python3-dev, python3-setuptools, libtool, libffi-dev, libssl-dev, autoconf, build-essential]
+parts:
+  alsa-mixin:
+    build-packages:
+    - libasound2-dev
+    plugin: dump
+    source: https://github.com/diddlesnaps/snapcraft-alsa.git
+    source-subdir: snapcraft-assets
+    stage-packages:
+    - libslang2
+    - libasound2
+    - libasound2-plugins
+    - libportaudio2
+    - pulseaudio
+    - alsa-utils
+  mycroft-precise:
+    after: [alsa-mixin]
+    plugin: python
+    python-version: python3
+    source: https://github.com/MycroftAI/mycroft-precise.git
+    source-type: git
+    source-branch: dev
+    override-pull: |
+        snapcraftctl pull
+        # Set version to setup.py version if current commit is the latest
+        # tag, otherwise append "-dev" to the version string.
+        set +e
+        if [[ git describe --exact-match --tags $(git log -n1 --pretty='%h') ]]
+        then
+            VERSION=$(/usr/bin/python3 setup.py --version)
+        else
+            VERSION="$(/usr/bin/python3 setup.py --version)-dev"
+        fi
+        set -e
+        snapcraftctl set-version $VERSION
+    build-packages:
+        - python-setuptools
+        - python3-pip
+        - curl
+        - libopenblas-dev
+        - python3-scipy
+        - cython
+        - libhdf5-dev
+        - python3-h5py
+        - portaudio19-dev
+        - swig
+        - libatlas-base-dev
+    stage-packages:
+        - libopenblas-base
+        - python3
+        - python3-scipy
+        - cython
+        - libhdf5-100
+        - libhdf5-cpp-100
+        - python3-h5py
+        - swig
+        - libatlas3-base
+  hey-mycroft:
+    plugin: dump
+    source: https://github.com/MycroftAI/precise-data/raw/models/hey-mycroft.tar.gz
+    source-type: tar
+    organize:
+        '*': /hey-mycroft/
+
+layout:
+  /usr/lib/$SNAPCRAFT_ARCH_TRIPLET/alsa-lib:
+    bind: $SNAP/usr/lib/$SNAPCRAFT_ARCH_TRIPLET/alsa-lib
+
+apps:
+  mycroft-precise:
+    plugs: [home]
+    command: bin/precise-engine
+  add-noise:
+    plugs: [home]
+    command: bin/precise-add-noise
+  calc-threshold:
+    plugs: [home]
+    command: bin/precise-calc-threshold
+  collect:
+    plugs: [home, audio-record, audio-playback]
+    command-chain:
+    - snap/command-chain/alsa-launch
+    command: bin/precise-collect
+  convert:
+    plugs: [home]
+    command: bin/precise-convert
+  eval:
+    plugs: [home]
+    command: bin/precise-eval
+  engine:
+    plugs: [home]
+    command: bin/precise-engine
+  graph:
+    plugs: [home]
+    command: bin/precise-graph
+  listen:
+    command-chain:
+    - snap/command-chain/alsa-launch
+    plugs: [home, audio-record, audio-playback]
+    command: bin/precise-listen
+  listen-pocketsphinx:
+    command-chain:
+    - snap/command-chain/alsa-launch
+    plugs: [home, audio-record, audio-playback]
+    command: bin/precise-listen-pocketsphinx
+  simulate:
+    plugs: [home]
+    command: bin/precise-simulate
+  test:
+    plugs: [home]
+    command: bin/precise-test
+  test-pocketsphinx:
+    plugs: [home]
+    command: bin/precise-test-pocketsphinx
+  train:
+    plugs: [home]
+    command: bin/precise-train
+  train-generated:
+    plugs: [home]
+    command: bin/precise-train-generated
+  train-incremental:
+    plugs: [home]
+    command: bin/precise-train-incremental
+  train-optimize:
+    plugs: [home]
+    command: bin/precise-train-optimize
+  train-sampled:
+    plugs: [home]
+    command: bin/precise-train-sampled