RE: [voiceinteraction] Open Voice Assistant Manifest

Hi Dirk,

These are all good points. I think the best way to track them would be to add them to the Open Voice docs repo issues, https://github.com/open-voice-interoperability/docs/issues, but I think we should discuss them in the Voice Interaction call first to make sure that the issues will be clear to the Open Voice group.

I think these would be good topics for the call tomorrow.

Debbie

 

From: Dirk Schnelle-Walka <dirk@switch-consulting.de> 
Sent: Monday, November 18, 2024 10:52
To: public-voiceinteraction@w3.org
Subject: Re: [voiceinteraction] Open Voice Assistant Manifest

 

Thank you Debbie,

this could be one of the things that we may link to in our architecture draft.

Currently, I find it a bit limited. At a first glance, I see

*         I am missing a clear identifier

*         supportedLayers and others seem to be restricted to written or spoken language, only

*         Only vague and not standardized means to describe the capabilities

*         serviceEndpoint is mandatory but there might be different means to access than by a URL, e.g. direct service call

*         No mentioning about authentication

Dirk

 

On 06/11/2024 19:17, Deborah Dahl wrote:

I put a link to more information about the Open Voice Assistant Manifest specification in the minutes today, but it didn't actually
have much information. Here's the actual specification.
https://github.com/open-voice-interoperability/docs/blob/main/specifications/AssistantManifest/0.9.0/AssistantManifestSpec.md
 
The purpose of this specification is to provide a structured way for an assistant like an IPA to describe itself and its
capabilities as part of the process of discovering assistants that could handle specific topics.