User Tools

Site Tools


eg-speech

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
eg-speech [2019/05/15 13:14]
waltminer
eg-speech [2019/11/05 02:03] (current)
waltminer Changed to winter time
Line 17: Line 17:
  
 ===== Meetings ===== ===== Meetings =====
-Meetings of the Speech EG are held every other Wednesday. Meeting time is 13:00 UTC. The upcoming schedule can be found below. ​+Meetings of the Speech EG are held every other Wednesday. Meeting time is 14:00 UTC. The upcoming schedule can be found below. ​
  
 Join Zoom Meeting\\ Join Zoom Meeting\\
Line 26: Line 26:
  
 ---------- ----------
 +
 +==== October 16, 2019 ====
 +Attendees: Walt, Scott, Jan-Simon, Michael, Li
 +
 +  * On alexa auto sdk v2.0 using an integrated platform build (kudos to Scott Murray)
 +  * Updates since Karlsruhe
 +     * Audio input and output both work on master according to Jan-Simon. ​ Raquel testing on halibut and we think it works there as well. 
 +     * Scott pushed rework to agl-service-voicehigh to remove hard dependencies on Alexa. ​
 +     * Scott reworking home screen changes from Naveen.
 +     * Need wake word solution from Amazon
 +     * Waiting on ICS for app code. 
 +
 +New: 
 +
 +
 +==== October 02, 2019 ====
 +Attendees: Walt, Jan-Simon, Kusakabe-san,​ Kurokawa-san
 +
 +  * On alexa auto sdk v2.0 using an integrated platform build (kudos to Scott Murray)
 +  * Integration Session in Karlsruhe:
 +     * Mic works, audio out works, but not at same time.
 +     * Fixes WIP for pipewire
 +     * agl-service-voiceagent-alexa needs fixes for smack/​multiuser
 +       * issues are known and resolved, but need to be integrate
 +       * AI: Jan-Simon/​Scott write-up steps in confluence
 +       * AI: Walt: generic icon for TTS-button
 +     * Steering wheel: we got first lin message but infrequent, need to research BCM (in-kernel scheduler for polling) and / or have the can-low-level listen (worst case configure BCM as well).
 +  * ICS-app: waiting for new binaries for AMM.  (aarch64 + x86-64)
 +
 +New: 
 +  * 
 +
 +==== September 18, 2019 ====
 +Attendees: Walt, Jan-Simon, Scott, Kusakabe
 +
 +   * Thierry still waiting for Naveen'​s patches that were discussed during the last call. 
 +     * 8/21: No update
 +     * 9/4: Thierry not available today. We do not think he has received anything from Naveen. ​
 +   ​* ​
 +   * Thierry needs a HH SDK that is stable and not a moving target. Will contact Jan-Simon to get the Halibut release folder created and the SDK populated. ​
 +     * 8/21: Done.
 +   * Soya-san has the new front end and is testing it with a number of new features and has a Microsemi Field Apps Engineer colleague in the Tokyo office to get some support with tuning the device. This will include features such as barge-in and better responsiveness.
 +     * 8/21: Mic being optimized (far-field/​near-field adaptation)
 +     * 9/4: Still in progress. Delayed by vacations and availability of the Microsemi FAE.  ​
 +   * Todo: check with new version of the Alexa SDK
 +     * Required: port to pipewire output (https://​jira.automotivelinux.org/​browse/​SPEC-2761)
 +     * Put https://​github.com/​alexa/​alexa-auto-sdk/​tree/​1.6/​platforms/​agl/​alexa-voiceagent-service build in recipe or improve build situation (currently container + external SDK is necessary) ​
 +
 +New: 
 +   * Planning an integration session the week after the Berlin F2F in Karlsruhe. Need Amazon and possibly ICS support. ​
 +
 +
 +==== September 4, 2019 ====
 +Attendees: Walt, Jan-Simon, Michael, Fulup, Scott
 +
 +   * Thierry still waiting for Naveen'​s patches that were discussed during the last call. 
 +     * 8/21: No update
 +     * 9/4: Thierry not available today. We do not think he has received anything from Naveen. ​
 +   ​* ​
 +   * Thierry needs a HH SDK that is stable and not a moving target. Will contact Jan-Simon to get the Halibut release folder created and the SDK populated. ​
 +     * 8/21: Done.
 +   * Soya-san has the new front end and is testing it with a number of new features and has a Microsemi Field Apps Engineer colleague in the Tokyo office to get some support with tuning the device. This will include features such as barge-in and better responsiveness.
 +     * 8/21: Mic being optimized (far-field/​near-field adaptation)
 +     * 9/4: Still in progress. Delayed by vacations and availability of the Microsemi FAE.  ​
 +   * Todo: check with new version of the Alexa SDK
 +     * Required: port to pipewire output (https://​jira.automotivelinux.org/​browse/​SPEC-2761)
 +     * Put https://​github.com/​alexa/​alexa-auto-sdk/​tree/​1.6/​platforms/​agl/​alexa-voiceagent-service build in recipe or improve build situation (currently container + external SDK is necessary) ​
 +
 +New: 
 +   * Planning an integration session the week after the Berlin F2F in Karlsruhe. Need Amazon and possibly ICS support. ​
 +
 +
 +==== August 21, 2019 ====
 +Attendees: Jan-Simon, Thierry, Michael
 +
 +   * Thierry still waiting for Naveen'​s patches that were discussed during the last call. 
 +     * 8/21: No update
 +   * Thierry needs a HH SDK that is stable and not a moving target. Will contact Jan-Simon to get the Halibut release folder created and the SDK populated. ​
 +     * 8/21: Done.
 +   * Soya-san has the new front end and is testing it with a number of new features and has a Microsemi Field Apps Engineer colleague in the Tokyo office to get some support with tuning the device. This will include features such as barge-in and better responsiveness.
 +     * 8/21: Mic being optimized (far-field/​near-field adaptation)
 +   * Todo: check with new version of the Alexa SDK
 +     * Required: port to pipewire output (https://​jira.automotivelinux.org/​browse/​SPEC-2761)
 +     * Put https://​github.com/​alexa/​alexa-auto-sdk/​tree/​1.6/​platforms/​agl/​alexa-voiceagent-service build in recipe or improve build situation (currently container + external SDK is necessary) ​
 +
 +New: 
 +   ​* ​
 +
 +
 +----------
 +
 +==== June 12, 2019 ====
 +Attendees: Walt, Thierry, Michael, Chris, Kusakabe
 +
 +   * Thierry still waiting for Naveen'​s patches that were discussed during the last call. 
 +   * Thierry needs a HH SDK that is stable and not a moving target. Will contact Jan-Simon to get the Halibut release folder created and the SDK populated. ​
 +   * Soya-san has the new front end and is testing it with a number of new features and has a Microsemi Field Apps Engineer colleague in the Tokyo office to get some support with tuning the device. This will include features such as barge-in and better responsiveness. ​
 +
  
 ==== May 29, 2019 ==== ==== May 29, 2019 ====
-Attendees: ​//Upcoming Meeting// ​+Attendees: ​Walt, Jan-Simon, Thierry, Naveen 
 + 
 +   * Naveen has two patches to commit to the IoT.bzh code to enable the open source agent on guppy. Thierry will integrate the patches then Jan-Simon can test on the green machine in Germany prior to ALS (preferably before end of June).
  
 +   * Email between Walt and Naveen
 +    - Plan to use Alexa demo during Automotive Linux Summit in July.  We are thinking we will reuse the CES demo and want to make sure we can do that **[Naveen]** Yes you can use that. This year we are mostly working on integrating our offline Alexa engine with Auto SDK. So, for next release of Auto SDK, which is tentatively planned for end of July, offline Alexa support will be added to AGL binding. This should add offline car control and local media capabilities to the Alexa binding and speech framework.
 +    - Plan for CES 2020.  Will Amazon be participating in CES 2020 and if so what are your ideas for that? **[Naveen]** Currently we don't have any plans for CES 2020. However, since we are updating the binding actively with every Auto SDK release, IOT.bzh will be in a position to host the Alexa demo from Linux foundation perspective.
 +    - Plan for F2F meeting in September. I am planning a September F2F meeting and need to choose a location. If Amazon plans to participate in CES 2020 I assume you will want to participate like you did last year in Santa Clara. ​ The only proposals I have on the table so far are for meetings in Europe. I know you have travel restrictions so if you plan to participate I will search harder for a US site for the meeting. **[Naveen]** Yes. I do have travel restrictions. However since our CES 2020 plans are not clear, please don't anchor F2F meeting based on Amazon'​s availability.
  
 ==== May 15, 2019 ==== ==== May 15, 2019 ====
Line 193: Line 297:
  
    * Documents now in Confluence ​    * Documents now in Confluence ​
-     * Architecture : https://​confluence.automotivelinux.org/​display/​SpeechArch/​Speech+EG+Architecture +     * Architecture : https://​confluence.automotivelinux.org/​display/​SPE/​Speech+EG+Architecture 
-     * CES Project ​ : https://​confluence.automotivelinux.org/​display/​SpeechArch/​Speech+EG%27s+CES+2019+Project+     * CES Project ​ : https://​confluence.automotivelinux.org/​display/​SPE/​Speech+EG%27s+CES+2019+Project
  
    * Reviewed questions and comments about latest Amazon proposal (v1.3). ​    * Reviewed questions and comments about latest Amazon proposal (v1.3). ​
Line 700: Line 804:
    * Walt will follow up with Amazon about their API. Schedule a follow up call for Tuesday, Feb 13     * Walt will follow up with Amazon about their API. Schedule a follow up call for Tuesday, Feb 13 
    * Question about how AGL handles app creation and installation    * Question about how AGL handles app creation and installation
-     * Developer guide for app developers is available at http://​docs.automotivelinux.org/​docs/devguides/en/dev/+   * Developer guide for app developers is available at https://​docs.automotivelinux.org/​docs/​en/​master/​devguides/
  
  
eg-speech.1557926086.txt.gz · Last modified: 2019/05/15 13:14 by waltminer