User Tools

Site Tools


eg-speech

This is an old revision of the document!


Speech Expert Group

Goals for Expert Group

  • Create a standardized set of speech recognition APIs that app developers can use regardless of underlying speech engine
  • Natural language or grammar tree based
  • On board or cloud based speech
  • Text to Speech API
  • Signal processing for noise reduction and echo cancellation (future roadmap)
  • Grammar development tools
  • Amazon has an open API that could be used as a starting point
  • Goal is to have a straw man by the AMM in Tokyo, Feb. 20-21 with work completed by ALS in June.
  • Next step is to set up a call with the engineers from all three companies who will be working on this.

Meetings

EG Speech Biweekly Meeting

Please join my meeting from your computer, tablet or smartphone.
https://global.gotomeeting.com/join/356562157

You can also dial in using your phone.
United States (Toll Free): 1 877 568 4106
United States: +1 (571) 317-3129

Access Code: 356-562-157

More phone numbers

Australia: +61 2 9087 3604 Austria: +43 7 2081 5427
Belgium: +32 28 93 7018 Canada (Toll Free): 1 888 455 1389
Canada: +1 (647) 497-9391 Denmark: +45 43 31 47 82
Finland: +358 923 17 0568 France: +33 170 950 594
Germany: +49 692 5736 7317 India (Toll Free): 18002669272
Ireland: +353 16 572 651 Italy: +39 0 291 29 46 30
Japan (Toll Free): 0 120 663 800 Korea, Republic of (Toll Free): 00798 14 207 4914
Netherlands: +31 207 941 377 New Zealand: +64 9 280 6302
Norway: +47 21 93 37 51 Spain: +34 932 75 2004
Sweden: +46 853 527 827 Switzerland: +41 225 4599 78
United Kingdom: +44 330 221 0088

Joining from a video-conferencing room or system?
Dial: 67.217.95.2##356562157
Cisco devices: 356562157@67.217.95.2

First GoToMeeting? Let's do a quick system check: https://link.gotomeeting.com/system-check

August 8, 2018

Attendees:

LF: Walt, Jan-Simon
Nuance: Christian, Paul Purcell, Mike C., Vince, Arijit
Amazon: Premal, Ankur, Naveen, Kamal, Alain
NTT Data MSE: Imamura
Denso Ten: Kusakabe
Microchip: Michael, Christian
IoT.bzh: Stephane, Fulup
Konsulko: Matt P., Matt R.
Myscript: Olivier, Etienne

Notes:

Nuance still discussing internally about releasing their API. Christian working with AGL App FW and working with writing an AGL Service layer in github (https://github.com/Nuance-Mobility/agl-speech-interface)

  • Not wired up to a speech or TTS engine, more of a loop back test.
  • Will send the link to Konsulko and IoT.bzh to review the API for suggestions.
    • Done.
  • List of AGL services available can be seen at https://git.automotivelinux.org/
  • Need to figure out consent and privacy issues with AGL Identity Agent.
    • No update.
  • How to manage grammar and natural language APIs and split between services and apps?
  • How to integrate cloud speech applications?
    • Example: “Find me the closest pizza place” is processed in the cloud and the location and name are returned to the ECU. How is this then transmitted to the POI and/or navi app?
    • Update 7/25: update for FF use 8-channel CSL usb dac, about to land in gerrit.
  • Starting a demo project internally led by Paul. End of Aug early Sep they plan to have a design document together internally and will be ready with any questions/issue. Would be a good idea to target the Sep F2F in Santa Clara to resolve issues with the design.
  • Information on IRC, mail list etc is available at Getting Started with AGL
  • Supported hardware can be found at AGL Distribution
  • 8/8
    • Arijit received the M3 hardware and was able to get it running. Building a “Hello World” sample application using Virtual Box and M3 hardware.

Amazon looking at releasing a possible API in June and starting to work with the AGL App FW.

  • Starting to look at AGL App FW binder implementation using audio HAL as a reference. Will work with IoT.bzh on how to put the configuration together.
    • Update 7/25: reviewing above draft, will share ideas/design to run multiple engines in parallel. No timeline, yet. Will review internally and present in next call.
  • Would like to put together an architecture picture based on the white board drawings from February AMM to see how the API fits into AGL overall.
  • 8/8
    • Naveen presented some use cases and an architecture diagram that Amazon has been working on internally. Received good feedback from the team. Naveen and his team will update their internal wiki and present again at the next meeting. Will look into getting the info onto the AGL wiki after that.

Microchip - AGL USB microphone front-end. Michael is working with MicroSemi on getting hardware available that is already available for Amazon Alexa. Would like to have a prototype available for the AMM in Dresden. Microchip plans to provide the HAL for the microphone.

  • Update 7/25:
    • Received hardware from MicroSemi, Alexa stack already working.
    • Hardware is mic+dsp connected to rpi running the stack
    • Plan: frontend should be connected over USB, integrated with 4a (hal) and interacting with the stack
      • Stack needs to pick-up conditioned signal (near/far/noise-cancelling) through alsa device
      • Michrochip will provide the hal for 4a
      • Michael: Interest to extend the API for beamforming, multiple seats, “1 channel per seat” ?
  • Question from Nucance about audio streaming:
    • esoundlib - do you need special calls to stream audio
    • Fulup: no, reply of 4a role request is the alsa device to write to
    • 4a-play /usr/share/4a/media/Happy_MBB_75.ogg (only script)

Action item:

  • Move the github repo into AGL git to foster collaboration - Done

July 25, 2018

Attendees:

LF: Walt, Jan-Simon
Nuance: Christian, Paul Purcell, Mike C., Vince, Arijit
Amazon: Premal, Ankur, Naveen, Kamal
NTT Data MSE: Imamura
Denso Ten: Kusakabe
Microchip: Michael, Christian
IoT.bzh: Stephane, Fulup
Konsulko: Matt P., Matt R.
Myscript: Olivier, Etienne

Notes:

Nuance still discussing internally about releasing their API. Christian working with AGL App FW and working with writing an AGL Service layer in github (https://github.com/Nuance-Mobility/agl-speech-interface)

  • Not wired up to a speech or TTS engine, more of a loop back test.
  • Will send the link to Konsulko and IoT.bzh to review the API for suggestions.
    • Done.
  • List of AGL services available can be seen at https://git.automotivelinux.org/
  • Need to figure out consent and privacy issues with AGL Identity Agent.
    • No update.
  • How to manage grammar and natural language APIs and split between services and apps?
  • How to integrate cloud speech applications?
    • Example: “Find me the closest pizza place” is processed in the cloud and the location and name are returned to the ECU. How is this then transmitted to the POI and/or navi app?
    • Update 7/25: update for FF use 8-channel CSL usb dac, about to land in gerrit.
  • Starting a demo project internally led by Paul. End of Aug early Sep they plan to have a design document together internally and will be ready with any questions/issue. Would be a good idea to target the Sep F2F in Santa Clara to resolve issues with the design.
  • Information on IRC, mail list etc is available at Getting Started with AGL
  • Supported hardware can be found at AGL Distribution

Amazon looking at releasing a possible API in June and starting to work with the AGL App FW.

  • Starting to look at AGL App FW binder implementation using audio HAL as a reference. Will work with IoT.bzh on how to put the configuration together.
    • Update 7/25: reviewing above draft, will share ideas/design to run multiple engines in parallel. No timeline, yet. Will review internally and present in next call.
  • Would like to put together an architecture picture based on the white board drawings from February AMM to see how the API fits into AGL overall.

Microchip - AGL USB microphone front-end. Michael is working with MicroSemi on getting hardware available that is already available for Amazon Alexa. Would like to have a prototype available for the AMM in Dresden. Microchip plans to provide the HAL for the microphone.

  • Update 7/25:
    • Received hardware from MicroSemi, Alexa stack already working.
    • Hardware is mic+dsp connected to rpi running the stack
    • Plan: frontend should be connected over USB, integrated with 4a (hal) and interacting with the stack
      • Stack needs to pick-up conditioned signal (near/far/noise-cancelling) through alsa device
      • Michrochip will provide the hal for 4a
      • Michael: Interest to extend the API for beamforming, multiple seats, “1 channel per seat” ?
  • Question from Nucance about audio streaming:
    • esoundlib - do you need special calls to stream audio
    • Fulup: no, reply of 4a role request is the alsa device to write to
    • 4a-play /usr/share/4a/media/Happy_MBB_75.ogg (only script)

Action item:

  • Move the github repo into AGL git to foster collaboration

July 11, 2018

Attendees:

LF: Walt, Jan-Simon
Voicebox:
Nuance: Christian, Paul Purcell, Mike C., Vince
Amazon: Premal, Ankur, Naveen
NTT Data MSE: Imamura
Denso Ten: Kusakabe
Microchip: Michael, Christian
IoT.bzh: Stephane, Fulup
Konsulko: Matt P., Matt R.

Notes:

Nuance still discussing internally about releasing their API. Christian working with AGL App FW and working with writing an AGL Service layer in github (https://github.com/Nuance-Mobility/agl-speech-interface)

  • Not wired up to a speech or TTS engine, more of a loop back test.
  • Will send the link to Konsulko and IoT.bzh to review the API for suggestions.
  • List of AGL services available can be seen at https://git.automotivelinux.org/
  • Need to figure out consent and privacy issues with AGL Identity Agent.
  • How to manage grammar and natural language APIs and split between services and apps?
  • How to integrate cloud speech applications?
    • Example: “Find me the closest pizza place” is processed in the cloud and the location and name are returned to the ECU. How is this then transmitted to the POI and/or navi app?
  • Starting a demo project internally led by Paul. End of Aug early Sep they plan to have a design document together internally and will be ready with any questions/issue. Would be a good idea to target the Sep F2F in Santa Clara to resolve issues with the design.
  • Information on IRC, mail list etc is available at Getting Started with AGL
  • Supported hardware can be found at AGL Distribution

Amazon looking at releasing a possible API in June and starting to work with the AGL App FW.

  • Starting to look at AGL App FW binder implementation using audio HAL as a reference. Will work with IoT.bzh on how to put the configuration together.
  • Would like to put together an architecture picture based on the white board drawings from February AMM to see how the API fits into AGL overall.

Microchip - AGL USB microphone front-end. Michael is working with MicroSemi on getting hardware available that is already available for Amazon Alexa. Would like to have a prototype available for the AMM in Dresden. Microchip plans to provide the HAL for the microphone.

Action item:

  • Move the github repo into AGL git to foster collaboration

June 27, 2018

Attendees:

LF: Walt, Jan-Simon
Voicebox:
Nuance: Christian, Mike C., Vince
Amazon: Premal, Ankur, Naveen
NTT Data MSE: Imamura
Denso Ten: Kusakabe
Microchip: Michael, Christian
IoT.bzh: Stephane, Fulup
Konsulko: Matt P., Matt R.

Notes:

Nuance still discussing internally about releasing their API. Christian working with AGL App FW and working with writing an AGL Service layer in github (https://github.com/Nuance-Mobility/agl-speech-interface)

  • Not wired up to a speech or TTS engine, more of a loop back test.
  • Will send the link to Konsulko and IoT.bzh to review the API for suggestions.
  • List of AGL services available can be seen at https://git.automotivelinux.org/
  • Need to figure out consent and privacy issues with AGL Identity Agent.
  • How to manage grammar and natural language APIs and split between services and apps?
  • How to integrate cloud speech applications?
    • Example: “Find me the closest pizza place” is processed in the cloud and the location and name are returned to the ECU. How is this then transmitted to the POI and/or navi app?

Amazon looking at releasing a possible API in June and starting to work with the AGL App FW.

  • Starting to look at AGL App FW binder implementation using audio HAL as a reference. Will work with IoT.bzh on how to put the configuration together.
  • Would like to put together an architecture picture based on the white board drawings from February AMM to see how the API fits into AGL overall.

Voicebox was acquired by Nuance so they will probably not be participating as a separate entity.

June 7, 2018

Attendees:

LF: Walt, Jan-Simon
Voicebox:
Nuance: Christian, Mike C., Vince
Amazon:
NTT Data MSE: Imamura
Denso Ten: Kusakabe
Microchip: Michael, Christian
IoT.bzh: Stephane, Fulup
Konsulko: Matt P., Matt R.

Notes:

Nuance still discussing internally about releasing their API. Christian working with AGL App FW and working with writing an AGL Service layer in github (https://github.com/Nuance-Mobility/agl-speech-interface)

  • Not wired up to a speech or TTS engine, more of a loop back test.
  • Will send the link to Konsulko and IoT.bzh to review the API for suggestions.
  • List of AGL services available can be seen at https://git.automotivelinux.org/
  • Need to figure out consent and privacy issues with AGL Identity Agent.
  • How to manage grammar and natural language APIs and split between services and apps?
  • How to integrate cloud speech applications?
    • Example: “Find me the closest pizza place” is processed in the cloud and the location and name are returned to the ECU. How is this then transmitted to the POI and/or navi app?

Amazon looking at releasing a possible API in June and starting to work with the AGL App FW.

  • No one joined.

Face-to-Face meeting planned for June 19 in Tokyo.

May 30, 2018

Attendees:

LF: Walt, Jan-Simon
Voicebox:
Nuance: Christian, Mike C., Vince
Amazon:
NTT Data MSE:
Denso Ten: Kusakabe
Microchip: Michael, Christian
IoT.bzh: Stephane
Qt Company:

Notes:

Nuance still discussing internally about releasing their API. Christian working with AGL App FW and working with writing an AGL Service layer in github (https://github.com/Nuance-Mobility/agl-speech-interface)

  • Not wired up to a speech or TTS engine, more of a loop back test.
  • Will send the link to Konsulko and IoT.bzh to review the API for suggestions.
  • List of AGL services available can be seen at https://git.automotivelinux.org/
  • Need to figure out consent and privacy issues with AGL Identity Agent.
  • How to manage grammar and natural language APIs and split between services and apps?
  • How to integrate cloud speech applications?
    • Example: “Find me the closest pizza place” is processed in the cloud and the location and name are returned to the ECU. How is this then transmitted to the POI and/or navi app?

Amazon looking at releasing a possible API in June and starting to work with the AGL App FW.

  • No one joined.

Video conference during the Lorient F2F meeting on June 7

Face-to-Face meeting planned for June 19 in Tokyo.


May 16, 2018

Attendees: LF: Walt
Voicebox:
Nuance: Christian
Amazon: Premal
NTT Data MSE: Imamura
Denso Ten: Kusakabe
Qt Company: Alistair

Notes:

Nuance still discussing internally about releasing their API. Christian working with AGL App FW and working with writing an AGL Service layer.

Amazon looking at releasing a possible API in June and starting to work with the AGL App FW.

Video conference during the Lorient F2F meeting on June 7

Face-to-Face meeting planned for June 19 in Tokyo.

Feb 14, 2018

Attendees:
LF: Walt, Dan
Voicebox: Andrew
Nuance: Christian, Mike
Amazon: John

Notes:

  • Amazon still internally discussing making their API available for AGL.
  • Amazon SDKs that are available publicly require an agreement with Amazon to access
    • Alexa Voice Service Device SDK
    • Alexa Skills Kit
  • Nuance interface still discussing internally. Christian made a presentation that he will send around with Nuance's ideas for the API.
  • Walt will use the minutes from these calls to lead an AMM session next week that updates the community on the EG's work.

Feb 6, 2018

Attendees:
LF: Walt, Jan-Simon
Voicebox: Andrew and Adam
Nuance: Christian, Mike
Amazon: None

Notes:

  • No word from Amazon on the availability of their API as well the link to what is publicly.
  • Nuance interface still discussing internally. Still waiting for Amazon proposal.
  • 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

Kick off Meeting Jan 29, 2018

Attendees: Walt, Jan-Simon, Mike Chachich, John Scumniotales, Andrew Fairly, Vince Iannotti, Christian Benien (attending AMM)

Agenda for kick off meeting

  • Review expert group goals as captured above
  • Proposal for speech recognition API or TTS API from Amazon?
  • Meeting schedule (biweekly? what time?)
  • Developer commitment from EG members.
  • Reviewed agenda and notes from CES
  • John said there is ongoing internal Amazon about releasing their code. There is already an open source version or public API version. Need to find out definitively what they are talking about as the release.
    • John will send a link to what is publicly available now
    • Should wrap up internal discussions end of this week. (Feb 2)
  • Michael may have something they can release. Will discuss internally once we see what Amazon has.
  • Attending AMM: Nuance: Christen, Amazon: Shitaro and Sanjay. VBT: TBD
  • Reserve time at AMM on Thursday (1 hour)
  • Follow up Feb 6
eg-speech.1533760460.txt.gz · Last modified: 2018/08/08 20:34 by waltminer