FedSearch - Federated network search engine

“Mm-hm,” “Uh-uh”: are non-lexical conversational sounds deal breakers for the ambient clinical documentation technology?
https://academic.oup.com/jamia/advance-article/doi/10.1093/jamia/ocad001/6998092

Think I can use this #JAMIA article to convince my 6 year old to stop with the Mm-hm and Uh-uh?
#Informatics #NLP #ASR

#jamia #informatics #nlp #ASR

Last updated 2 years ago

Original post

Geekmaster 👽:system76: · @Geekmaster

121 followers · 931 posts · Server ioc.exchange

⚠️ #MIcrosoftServiceHealth #Advisory MO497128: For everyone who lost the use of their #MicrostfOffice desktop apps today, it's because of an issue that Microsoft is dealing with, related directly to #Defender #AttackSurfaceReduction, or #ASR rules. Specifically: "Block Win32 API calls from Office Macros" with ID 92e97fa1-2edf-4476-bdd6-9dd0b4dddc7b.

#Hotfix:
Admins can put the ASR rule into #Audit Mode to avoid further impact. Please note that you may need to re-enable the rule once the issue has been fully resolved. This can be done through one of the following methods:

- Using Powershell: Add-MpPreference -AttackSurfaceReductionRules_Ids 92e97fa1-2edf-4476-bdd6-9dd0b4dddc7b -AttackSurfaceReductionRules_Actions AuditMode

- Using Intune: https://learn.microsoft.com/en-us/microsoft-365/security/defender-endpoint/enable-attack-surface-reduction?view=o365-worldwide#mem

- Using Group Policy: https://learn.microsoft.com/en-us/microsoft-365/security/defender-endpoint/enable-attack-surface-reduction?view=o365-worldwide#group-policy

#TheMoreYouKnow #SysAdmins

#microsoftservicehealth #advisory #microstfoffice #defender #attacksurfacereduction #ASR #hotfix #audit #themoreyouknow #sysadmins

Last updated 2 years ago

Original post

Giorgio Robino · @solyarisoftware

36 followers · 29 posts · Server sigmoid.social

Open media

I fully agree with
David Shapiro (@dshap_automator)
when he says:

Ehi
@Microsoft
, why don't you use #ChatGPT revamping dead #Cortana?!

Ingredients:
#LLM: a GPT3.5 model
#ASR: Openai Whisper
#TTS: A 'contextual voice model' like Roger

Don't worry about Bing!

Video:
https://youtube.com/watch?v=1gCb_48wP9A

#chatgpt #cortana #llm #ASR #tts

Last updated 2 years ago

Original post

Geekmaster 👽 · @Geekmaster

91 followers · 723 posts · Server ioc.exchange

IFCR - Pass-the-Challenge: Defeating Windows Defender Credential Guard

This is why you configure #ASR rules in #Defender and also switch to #Kerberos rather than #NTLM #SecureYourEnterprise before you get #hacked https://research.ifcr.dk/pass-the-challenge-defeating-windows-defender-credential-guard-31a892eee22

#ASR #defender #kerberos #ntlm #secureyourenterprise #hacked

Last updated 2 years ago

Original post

Stavroula Sokoli · @stavroula

78 followers · 40 posts · Server mastodon.online

#Skype uses #ASR and #NLP to translate speech.
#AI is then used on top of that, sampling spoken words to shape the sound of the #voicetranslation. The natural voice processing is done in real-time to prevent people from being able to misuse your voice.
Available "next year"

https://www.youtube.com/watch?v=qrTZ2IQpwi0&t=2s

#voicetranslation #ai #nlp #ASR #skype

Last updated 2 years ago

Original post

Joshua McNeill · @joshisanonymous

97 followers · 191 posts · Server h4.io

@ElenLeFoll I really can't wait for the day when #ASR can handle even quasi-#codeswitching. I wanna be able to tell my phone in French to play music by an artist with an English name!

#ASR #codeswitching

Last updated 2 years ago

Original post

Vaclav Hanzl · @vaclavh

15 followers · 24 posts · Server sigmoid.social

@naturepoker Just googling related articles, yes, interest in phage therapy does not mark one as charlatan anymore. (And even charlatans do useful things sometimes - in speech processing my favorite is von Kempelen who faked a chess machine in half a year and then spent decades on final detail making it to say "check" - doing in fact serious research and founding this field. Most people in #ASR heard both stories but do not realize that it was the same person :) )

#ASR

Last updated 2 years ago

Original post

Vaclav Hanzl · @vaclavh

6 followers · 3 posts · Server sigmoid.social

#introduction
I did #ASR and #DSP from early days when recording of 3 words was daring test of limits for our self-soldered computers till present #cloud #GPU #AI, going through academia, startup, big tech company and independent work. At every career fork I also consider #genomics, being fascinated by similarities in computer and living beings architectures ever since I realized that #RNA is both software and hardware. Also #neuroscience also shares many mysteries with my digital world.

#Introduction #ASR #dsp #cloud #gpu #ai #genomics #RNA #neuroscience

Last updated 2 years ago

Original post

Mike Harms · @EvolBiochemist

179 followers · 68 posts · Server ecoevo.social

@ShropshireJD We've been using muscle5 for protein #phylogenetics and #ASR. Mind blowing results, at least for our systems (mostly vertebrate proteins).

#phylogenetics #ASR

Last updated 2 years ago

Original post

Matt Coler · @Mattcoler

328 followers · 53 posts · Server fediscience.org

Hello Mastodon! Here's my belated #introduction. I am an Associate Professor of Language & Technology and the Director of the MSc Voice Technology at the University of Groningen. #OpenAcess Ambassador.

Interests: #SpeechTechnology, #VoiceTechnology, voice #synthesis, #speech #recognition #ASR, #speechAI, #multisensory perception, #audition, #soundscapes, #SituatedCognition, music, Andean languages, #Aymara, #Frisian

#Frisian #aymara #SituatedCognition #soundscapes #audition #multisensory #speechAI #ASR #recognition #speech #synthesis #VoiceTechnology #SpeechTechnology #OpenAcess #introduction

Last updated 2 years ago

Original post

pupazzaro · @pupazzaro

35 followers · 11 posts · Server mastodon.uno

Open media

Camelia Caesar, nomen omen. #asroma #asromawomen #daje #footballart #asr

#ASR #footballart #daje #asromawomen #asroma

Last updated 2 years ago

Original post

KuboF Hromoslav · @kubofhromoslav

50 followers · 871 posts · Server esperanto.masto.host

RT @alphacep@twitter.com

Esperanto contributors doubled the amount of speech in Common Voice 9. We retrained Vosk Esperanto model with new data. WER improved from 8.28% on CV Test to 7.24%.
Thanks to everyone involved and @stergro@twitter.com

https://alphacephei.com/vosk/models/vosk-model-small-eo-0.42.zip

#esperanto #vosk #speechrecognition #asr

🐦🔗: https://twitter.com/alphacep/status/1538817183270289410

#esperanto #Vosk #speechrecognition #ASR

Last updated 2 years ago

Original post