Anybody experience with Alexa-Skills ?

[Rollo62 595]

Hi there,

 

I'm looking for some communication with the Alexa, and a custom skill.

The idea is that Alexa uses a AWS Lambda function, and may control a web-service to activate some stuff.

 

Thats usually the easy way, via Alexa Intents:

A.) Normal command flow

Quote

 

User speak ==> Alexa recognize intent ==> AWS new session ==> AWS Lambda controls/request something from web-service

...
... processing delay (1-2 sec.) ...
...

Web-service delivers data ==> AWS Lambda active session callback prepare speech-output ==> Alexa's voice outputs the answer

 

 

 

So far so good, the processing delay is usually 1-2 sec., so I still would call it real-time.

B.) Permanent command flow
What I need is to use unintended data transmissions from web-service.

 

I could see to ways to achive that:

 

1. Register some permanent callback:

Quote

Web-service delivers data ==> AWS Lambda permanent callback prepare speech-output ==> Alexa's voice outputs the answer

2. Keep the AWS session open

Quote

 

User command ==> Alexa open new, active session with callback

...
... keep session active, waiting for response (could be longer, like 30min, 60min, 2h, ... ) ?
...

Web-service delivers data ==> AWS Lambda active session callback prepare speech-output ==> Alexa's voice outputs the answer
==> Keep active session callback open, until user commands to stop it.

 

 

 

Either way: Alexa seems to be a one-way, as A.), I have not found too many info about using Alexa as a kind of TTS-output.
If 1.) or 2.) is possible at all, I cannot say, maybe there is even another option I don't see yet.

Like triggering any voice output by web-hook, or the like, that would be perfect.

 

Of coarse the application shall use Delphi/FMX to control finally, but the loophole seems to be the limited Alexa environment.

Maybe somebody has a helpful hint, or a good solution, to push me in the right direction ?

 

 

 

 

 

 

 

Edited September 29, 2020 by Rollo62