rhasspy-asr-pocketsphinx-hermes 0.1.4

No project description provided

Rhasspy ASR Pocketsphinx Hermes MQTT Service

Implements hermes/asr functionality from Hermes protocol using rhasspy-asr-pocketsphinx.

Running With Docker

docker run -it rhasspy/rhasspy-asr-pocketsphinx-hermes:<VERSION> <ARGS>

Building From Source

Make sure you have the required dependencies:

sudo apt-get update
sudo apt-get install build-essential

Clone the repository and create the virtual environment:

git clone https://github.com/rhasspy/rhasspy-asr-pocketsphinx-hermes.git
cd rhasspy-asr-pocketsphinx-hermes
make venv

Run the bin/rhasspy-asr-pocketsphinx-hermes script to access the command-line interface:

bin/rhasspy-asr-pocketsphinx-hermes --help

To re-train, you will also need to download pre-built binaries for mitlm and phonetisaurus, or build these programs yourself. The estimate-ngram and phonetisaurus-apply commands must be in your PATH.

Building the Debian Package

Follow the instructions to build from source, then run:

source .venv/bin/activate
make debian

If successful, you'll find a .deb file in the dist directory that can be installed with apt.

Building the Docker Image

Follow the instructions to build from source, then run:

source .venv/bin/activate
make docker

This will create a Docker image tagged rhasspy/rhasspy-asr-pocketsphinx-hermes:<VERSION> where VERSION comes from the file of the same name in the source root directory.

NOTE: If you add things to the Docker image, make sure to whitelist them in .dockerignore.

Command-Line Options

usage: rhasspy-asr-pocketsphinx-hermes [-h] --acoustic-model ACOUSTIC_MODEL
                                       --dictionary DICTIONARY
                                       --language-model LANGUAGE_MODEL
                                       [--mllr-matrix MLLR_MATRIX]
                                       [--base-dictionary BASE_DICTIONARY]
                                       [--reload RELOAD] [--host HOST]
                                       [--port PORT] [--siteId SITEID]
                                       [--stdin-files] [--debug]

optional arguments:
  -h, --help            show this help message and exit
  --acoustic-model ACOUSTIC_MODEL
                        Path to Pocketsphinx acoustic model directory (hmm)
  --dictionary DICTIONARY
                        Path to pronunciation dictionary file
  --language-model LANGUAGE_MODEL
                        Path to ARPA language model file
  --mllr-matrix MLLR_MATRIX
                        Path to tuned MLLR matrix file
  --base-dictionary BASE_DICTIONARY
                        Path(s) to base pronunciation dictionary file(s)
  --reload RELOAD       Poll dictionary/language model for given number of
                        seconds and automatically reload when changed
  --host HOST           MQTT host (default: localhost)
  --port PORT           MQTT port (default: 1883)
  --siteId SITEID       Hermes siteId(s) to listen for (default: all)
  --stdin-files         Read WAV file paths from stdin
  --debug               Print DEBUG messages to the console