I have an automation that triggers speech output on Alexa when certain actions happen.
However, I don’t really understand the difference between the two available Alexa integrations.
Currently, I have installed the custom “Alexa Media Player,” but after a few weeks I already started having connection issues. Is the other integration more reliable? Or is this something related to Amazon itself, meaning that all Alexa integrations will have similar problems?
I’m not sure which integration is better for handling speech output.
Question 1: Which integration is best for speech output on Alexa?
Question 2: After the update, I was able to reconnect to the Amazon cloud, but my automation no longer works.
Do actions often change after updates? What can I do to avoid having to check or modify all my automations after every update?
At the moment, I receive this error message. Ideally, I would like the system to be as stable as possible in the future.
(i know that we have always to update automation but i just want to try to make it not every month )
Whilst the Alexa Devices (AD) integration is the official HA integration for this, I’m using the HACS Alexa Media Player (AMP) integration. This is for two reasons:
The AMP integration has more features - although the AD integration will catch up in time.
if there is an issue with AD or AMP (due to some change that Amazon have made) then a fix for AMP is likely to be available sooner. This is because, HACS updates can be released at any time but updates for core integrations only get released once a week.
While @templeton_nash makes valid points, allow me to expand on the topic -
AMP uses an unofficial API via web access. AD uses the official API, the same one the Alexa app uses.
While AMP used to have an edge on being updated sooner due to HACS availability, this is no longer the case as the developers of AD have created an alpha/beta test version available in HACS using a custom repository.
The developers of AD have also previously stated that as the integration matures, becoming more stable and fully featured, AMP will be phased out.
Where did you see that AMP will be discontinued?
Which features are still missing in AD?
Does using the official API help reduce bugs?
I imagine the API changes less often than the web access, which is probably updated frequently. Does the AD version allow sending voice text (Alexa announcements)?
Actually, I installed AMP because I couldn’t find the YAML command for a voice announcement
I probably just didn’t look carefully…
I saw it here, on the community forums. The owner of the aioamazondevices branch, Chemelli, was the one who said it.
Currently AD is in active development. They just introduced a http wrapper (needed to add media_player and some sensors) into the integration. They are currently refactoring the code database. Announce and speak entities are working well. Do not disturb boolean is working well. There are rate limits on additional sensors, like temperature and motion sensors. They have been working to smooth those out. On their roadmap, media_player entities and connected devices (like bluetooth, zigbee, etc) are features they will be adding later.
The reason I brought up the API differences are because Amazon is way more likely to block access from an unofficial API.
Mind you, the current core version won’t work. There is an issue with logging in due to a change Amazon made. There is an active alpha test version in HACS that fixes the issue and does work.
So it might be better to wait for the final release of the official AD version… Right now, AMP seems to have more features. I understand now why it’s not the other way around and why AMP isn’t directly integrated into Home Assistant.
)
