Michael Auli (@michaelauli) 's Twitter Profile
Michael Auli

@michaelauli

Research Scientist @FacebookAI

ID: 1319131066259521536

linkhttps://michaelauli.github.io/ calendar_today22-10-2020 04:18:59

85 Tweet

2,2K Followers

154 Following

Michael Auli (@michaelauli) 's Twitter Profile Photo

I am looking for an autumn/winter 2022 PhD intern interested in speech and/or self supervised learning. Please DM me if you are interested with your CV.

Michael Auli (@michaelauli) 's Twitter Profile Photo

data2vec is a long talk at this year's ICML Conference. Congratulations to @ZloiAlexei Wei-Ning Hsu Qiantong Xu Arun Babu Jiatao Gu! Updated paper: arxiv.org/abs/2202.03555 with new ablations showing that contextualized target representations work very well.

Michael Auli (@michaelauli) 's Twitter Profile Photo

I will be at Interspeech next week! Looking forward to attending my first in-person conference for a long time and talking to people about our papers (XLS-R, unsupervised TTS, zero-shot phone recognition, more efficient wav2vec, etc.)

joao carreira (@joaocarreira) 's Twitter Profile Photo

Join us for the Computer Perception virtual workshop at ECCV this Sunday (computerperception.github.io). We will have keynotes from a wonderful set of speakers Olga Russakovsky, Aude Oliva, Matt Botvinick, Michael Auli, Daniel Yamins, Jitendra MALIK + talk about how to test perception.

Jiatao Gu (@thoma_gu) 's Twitter Profile Photo

We (Maha Elbayad Michael Auli Edouard Grave) have also worked on a very similar approach to adaptively change the depth of decoding for MT date back to ICLR2020 openreview.net/forum?id=SJg7K…

kyutai (@kyutai_labs) 's Twitter Profile Photo

Our founding team is covering many AI fields from vision, with Patrick Pérez and Hervé Jégou (Hervé Jegou) to LLMs with Edouard Grave (Edouard Grave), audio with Neil Zeghidour (Neil Zeghidour) and Alexandre Défossez (Alexandre Défossez) and infra with Laurent Mazaré (Laurent Mazare).

Our founding team is covering many AI fields from vision, with Patrick Pérez and Hervé Jégou (<a href="/hjegou/">Hervé Jegou</a>) to LLMs with Edouard Grave (<a href="/EXGRV/">Edouard Grave</a>), audio with Neil Zeghidour (<a href="/neilzegh/">Neil Zeghidour</a>) and Alexandre Défossez (<a href="/honualx/">Alexandre Défossez</a>) and infra with Laurent Mazaré (<a href="/lmazare/">Laurent Mazare</a>).
Michael Auli (@michaelauli) 's Twitter Profile Photo

We are releasing MMS Zero-shot: a model to transcribe the speech of almost any language using only a small amount of unlabeled text in the new language. Paper: arxiv.org/abs/2407.17852 Demo: huggingface.co/spaces/mms-met… Code/model: github.com/facebookresear…