Michael I Mandel

We’re thrilled to see our advanced ML models and EMG hardware — that transform neural signals controlling muscles at the wrist into commands that seamlessly drive computer interactions — appearing in the latest edition of @Nature. Read the story: nature.com/articles/s4158… Find more details on this work and the models on @github: github.com/facebookresear…

1/ We just open-sourced two large wrist electromyography (EMG) datasets - one towards typing without a keyboard and the other for predicting hand poses - with baselines. We believe these will help advance research into making high bandwidth non-invasive neuromotor interfaces a reality!

You might think speech recognition is "solved" with models such as @OpenAI’s Whisper, but it's not. Natural conversations with distant microphones still lack effective solutions. To illustrate, on our newly released NOTSOFAR meeting benchmark, Whisper large-v3 with head-mounted mics achieves 9.3% WER (word-error-rate), yet on audio from a distant mic it climbs to 37.4% WER. The culprits are reverberation, noise, and overlapping speech, which interfere with the source signal. What's the missing ingredient? We believe it's datasets. The problem is not amenable to web scraping. Benchmarking datasets are scarce given their complex collection process. Microphone arrays, useful for speech separation, are rarely featured in labeled datasets, necessitating simulation to teach neural networks to utilize such arrays. To bridge the gap our team at @Microsoft has released a benchmarking dataset of 280 recorded meetings, and a 1000-hour simulated training set synthesized for real-world generalization. Join our challenge "NOTSOFAR: Distant Meeting Transcription with a Single Device", part of CHiME-8, to explore these resources and advance the field. Details and registration: aka.ms/chime8 Code and datasets: aka.ms/notsofar

1/7 For the past decade, our team at Meta Reality Labs (previously CTRL-labs) has been dedicated to developing a neuromotor interface. Our goal is to address the Human Computer Interaction challenge of providing effortless, intuitive, and efficient input to computers.

Teaching DSP this Fall, so I'm relaunching my YouTube video series. If you/your colleagues are teaching DSP this year, your students might find useful. Pls LMK any errors, feedback, & requests for future videos! Course info here: creating-at-a-distance.com/applied-dsp/ youtube.com/playlist?list=…

Please check!

@asterix77 and I gave a talk at @Mila_Quebec covering some of our work at @RealityLabs towards building non-invasive neural interfaces using electromyography (EMG): #h.kk6bjvwshukm" target="_blank" rel="nofollow noopener">sites.google.com/lisa.iro.umont… Full talk: bluejeans.com/playback/s/JCb…

Want to know what we're up to in the neural interfaces group at @RealityLabs? -> Two members of the team I'm on recently gave a talk at @Mila_Quebec, check out our work on non-invasive neural interfaces: bluejeans.com/playback/s/JCb…

@mclduk @sarabssethi @Feral5Creative

@mclduk @sarabssethi We have lots of recordings from Alaska's north slope with some anthrophony. I think it's mostly cars and stationary oil machinery, but let me check.

Does anyone have eco-acoustic data from cold regions with snowmobiles / snow-scooters in the background? If so please reach out!

🧵 The #ismir2021 conference is underway! Here’s what’s happening with #WiMIR and Diversity & Inclusion around the conference... @ismir2021 #WomenInTech #WomenInStem #DiversityAndInclusion 1/ ismir2021.ismir.net

@PigEpidermis Good luck!

@peteskomoroch @TlkngMchns is launching season 7 soon

What are the most popular machine learning podcasts right now?

@colinraffel Transformers for publication, tfidf for production...

Come join myself and @asterix77 as @eva_zangerle rotates off this super interesting and important committee.

While @ismir2020 has come to a close, the sixth round of #WiMIR mentoring is just beginning! GET a mentor here! bit.ly/31ewCMP Sign up to be a mentor here! bit.ly/351yXfq Submissions close on Dec. 15, with matches announced in Jan 2021. #WomenInSTEM 1/3

Full #chimechallenge software baseline released in Kaldi. Instructions at chimechallenge.github.io/chime6/

Announcing the CHiME-6 Speech Separation and Recognition Challenge: spandh.dcs.shef.ac.uk/chime_challeng… Track 1: (repeat of CHiME-5) multichannel, multi-device speech recognition at dinner parties Track 2: same, but with multichannel, multi-device speaker diarization first

Keynote speakers announced for #chimeworkshop 2020: Paola García (Johns Hopkins University) and Dong Yu (Tencent AI Lab)

Also announcing the CHiME 2020 workshop on Speech Processing in Everyday Environments at Universitat Pompeu Fabra, Barcelona, May 4, 2020 (satellite to ICASSP 2020) spandh.dcs.shef.ac.uk/chime_workshop…

@PigEpidermis But you can stack them so much better!