Skip to main content

Library for creating voice assistant skills for the hermes skill server (hss-server)

Project description

HSS - Skill

Library for creating skills based on the Hermes Skill Server.

Installation

Simply use pip:

(hss) pi@ceres:~/development/myskill $ pip3 install hss_skill

Overview

The hss_skill package contains tools for fast and easy development of skills for the Hermes Skill Server. The goal is to let skill developers only care about their own skill implementation, while the internal stuff (communication with the skill-server, reading configuration, etc) is provided out-of-the-box by the hss_skill package.

The package provides a base class for skills BaseSkill which does all the incovenient stuff, like communication with the skill server, reading configuration file etc.

Abstract methods

When developing skills, a subclass of BaseSkill must implement two abstract methods, both of which are coroutines:

async def get_intentlist()

The implementation shall return a list of intents handled by your skill. This will be used in the hss-server to detect/avoid duplicate intent registration.

async def handle(request, session_id, site_id, intent_name, slots)

The implementation should add program logic here to perform actual intent handling. Usually, the parameters intent_name and slots might be sufficient, however the full original intent is provided in the request parameter, and session_id and site_id can be used to do session- and site-based intent handling.

The implementation of this method should usually return with the execution of either BaseSkill.answer or BaseSkill.followup.

Convenience methods

In addition, BaseSkill provides several methods which aid in skill development.

def answer(session_id, site_id, response_message, lang)

The answer-method should be called after the intent has been fully handled. This method also allows to send a response-text, which will then be forwarded to the TTS the your voice assistant.

The parameters session_id and site_id should be the ones provided by handle, while the text parameter shall be the text which shall be asked by the voice assistant.

def followup(session_id, site_id, question, lang, intent_filter = None)

The followup -method should be called when the skill does not yet want to finish handling, but instead needs to ask for additional input. The question-text will be forwarded to the TTS of the voice assistant. In addition, a filter for intents (array of strings) can be given (see hermes protocol docs).

The parameters session_id and site_id should be the ones provided by handle, while the question parameter shall be the text which shall be asked by the voice assistant.

async def say(text, siteId = None, lang = None)

The say coroutine can be used to trigger the voice assistant to say a given text using its TTS. There is no further session- or intent handling involved. Since say is a coroutine, it must be await-ed.

async def ask(text, siteId = None, lang = None, intent_filter = None)

The ask coroutine can be used to start a new session. This will usually cause the voice assistant to speak the provided text using its TTS, and then listen for intents. Recognized intents may then be processed again.

Optionally, an intent_filter (array of strings) can be given which will be forwarded to the voice assistant (see hermes protocol docs).

Since ask is a coroutine, it must be await-ed.

main.py

Skills must provide the file main.py, which is the file the skill-server is going to run. This file should create an instance of your skill class, and then call the run method of the skill.

Configuration

hss_skill automatically read a configuration file config.ini if it is present in your skills root-folder. The configuration will be provided to the skill-class via self.config.

Dependencies

Further dependencies can be defined in the file requirements.txt which should at least contain the dependency to hss_skill.

Example

A minimum example of using hss_skill. The folder contents might look like:

  • main.py
  • myskill.py
  • config.ini
  • requirements.txt

main.py

import myskill

if __name__ == "__main__":
    skill = myskill.MoodSkill()
    skill.run()

myskill.py

from hss_skill import hss

class MoodSkill(hss.BaseSkill):
    def __init__(self):
        super().__init__()   # important, call super's constructor

    async def get_intentlist(self):
        return ["howAreYou"]

    async def handle(self, request, session_id, site_id, intent_name, slots):
        return self.answer(session_id, site_id, "Thanks, I am fine")

requirements.txt

hss_skill>=0.1.2

Timers

The BaseSkill class provides a convenience method for setting up timers, which will execute a given callback function after a given timeout. This might be useful if the skill wants to trigger actions on its own at a given time.

Currently, as a limitation, only one timer can be active at a time. This will most likely change in the future, so that an arbitrary number of timers can be scheduled.

For this, two coroutines are provided:

async def timer(timeout, callback, user = None, reschedule = False)

Schedules a new timer. timeout shall be int and denote the numer of seconds until the provided coroutine callback is executed. If user is given, it will be passed to callback upon execution.

If a timer is already running, new scheduling will fail unless True is given for reschedule. In the latter case, the previous timer will be cancelled before a new timer is scheduled.

async def cancel_timer(strict = True)

Cancels an existing timer. If True is given for strict, an error message will be printed when cancel_timer is called but no timer is running.

Example

    async def handle(self, request, session_id, site_id, intent_name, slots):

        ... # skill handling code

        # schedule timer in 10 seconds

        await self.timer(10, self.do_timer, "Can I ask you a question?", reschedule = True)

        # finish intent handling

        return self.answer(session_id, site_id, response_message)

    async def do_timer(self, text):

        # ask a question

        await self.ask(text, siteId = "default", intent_filter = ["s710:confirm", "s710:reject"])

Skill installation

Please refer to Hermes Skill Server.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

hss_skill-0.3.0.tar.gz (7.7 kB view details)

Uploaded Source

Built Distribution

hss_skill-0.3.0-py3-none-any.whl (8.7 kB view details)

Uploaded Python 3

File details

Details for the file hss_skill-0.3.0.tar.gz.

File metadata

  • Download URL: hss_skill-0.3.0.tar.gz
  • Upload date:
  • Size: 7.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.23.0 setuptools/46.1.3 requests-toolbelt/0.9.1 tqdm/4.46.0 CPython/3.7.2

File hashes

Hashes for hss_skill-0.3.0.tar.gz
Algorithm Hash digest
SHA256 087e61cf7f3736303c48a9c987c84506992870e131abf57d2a1ae484fb6cff3f
MD5 bb0d5cd7d728a75659407c1da009795e
BLAKE2b-256 10e411d802d1e713695d92706b23a4269d30ee2ac092b5d2a558fff06de0422a

See more details on using hashes here.

File details

Details for the file hss_skill-0.3.0-py3-none-any.whl.

File metadata

  • Download URL: hss_skill-0.3.0-py3-none-any.whl
  • Upload date:
  • Size: 8.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.23.0 setuptools/46.1.3 requests-toolbelt/0.9.1 tqdm/4.46.0 CPython/3.7.2

File hashes

Hashes for hss_skill-0.3.0-py3-none-any.whl
Algorithm Hash digest
SHA256 26045c21a5f419c5e80e0af04f49a7867b18e1a108a98f5c0da1b1beebdbf321
MD5 1a125de58362a12595f800b178dca2a5
BLAKE2b-256 6e4a430c7280042fbb797e84d57845b40dfde7605fda738c53879211e40309cb

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page