Mirco Ravanelli
@mirco_ravanelli
Followers
4K
Following
2K
Media
96
Statuses
589
Deep learning for Conversational AI. Creator of SpeechBrain.
Montreal
Joined December 2017
π’ Join our Conversational AI Reading Group! π
Thursday, Oct 23rd | 11 AM - 12 PM EST π Speaker: Joan SerrΓ @serrjoa - Sony AI π Topic: "Supervised contrastive learning from weakly-labeled audio segments for musical version matching" π Details: ( https://t.co/oxEZtla7O1)
0
2
4
π’This week, our Conversational AI Reading Group is excited to have Jinyu Li from Microsoft. Please note: This weekβs session will start one hour later than usual, at 12:00 PM instead of 11:00 AM. π
Thursday, Oct 16th | 12:00 - 13:00 EST π Topic: The development of spoken LM
0
1
2
Tomorrow (Sept 25, 11:00β12:00 EST), our #ConversationalAI Reading Group hosts @Themos Stafylakis (Athens Univ. & Omilia). Talk: Advances in Speaker Recognition: Pruning, Deepfake Detection & Learning w/o Temporal Labels Info: https://t.co/idkmKuAYxi
#AI #SpeechProcessing
0
0
3
Our Conversational AI Reading Group (led by @MousaviPooneh) resumes tomorrow! Sept 18 | 11:00 AM EST Amazing speakers, great talks, open to all. Perfect for anyone in #DeepLearning & #ConversationalAI More info: https://t.co/tmU6I4G96W
#AI #MachineLearning #OpenScience
0
1
9
π’ Our Conversational AI Reading Group is back! Join the first Fall 2025 session! π€ π
Thursday, Sept 18 | 11 AMβ12 PM EST π Speaker: Pooneh Mousavi (Mila) @MousaviPooneh π Topic: βDiscrete Audio Tokens: More Than a Survey!β π Details:
0
2
5
π’ Presenting our paper βLiSTEN: Learning Soft Token Embeddings for Neural Audio LLMsβ β an interpretable fine-tuning method for spoken language understanding. π Wed, Aug 20 | 08:30β10:30 π A11-P2B-03 Hope to see you there! π https://t.co/GEhsFFqeLy
@ISCAInterspeech
1
2
6
Our pick of the week by @beomseok_lee_: "ALAS: Measuring Latent Speech-Text Alignment For Spoken Language Understanding In Multimodal LLMs" by Pooneh Mousavi, @yingzhi_wang, @mirco_ravanelli, and @CemSubakan (2025) https://t.co/8UsRFvV83N
#SLU #speech #multimodal #LLM
arxiv.org
Large Language Models (LLMs) are increasingly used in Spoken Language Understanding (SLU), where effective multimodal learning depends on the alignment between audio and text. Despite various...
Speech-language models show promise in multimodal tasksβbut how well are speech & text actually aligned? π€ This paper https://t.co/B9z1j7L4IO proposes a new metric to measure layer-wise correlation between the two, with a focus on SLU tasks. ππ£οΈπ
0
7
11
π We're excited to announce our latest work: "Discrete Audio Tokens: More Than a Survey!" It presents a comprehensive survey and benchmark of audio tokenizers across speech, music, and general audio. preprint: https://t.co/QogseMhEGH website:
1
14
34
π’ Join our Conversational AI Reading Group! π
Thursday, May 29th | 11 AM - 12 PM EST π Speaker: Yossi Adi @adiyossLC π Topic: "On The Landscape of Spoken Language Models" π Details: ( https://t.co/oxEZtl9zYt)
0
2
10
IEEE MLSP 2025: Deadline extended! The submission deadline is now May 6. - New papers welcome - Revisions accepted https://t.co/S0Xvsm8gfA Don't miss this chance to contribute! #MLSP2025 #MachineLearning #DeepLearning #SignalProcessing #IEEE #CallForPapers #Research
0
2
7
Deadline Extended !!! The paper submission deadline for #IEEE MLSP 2025 is now April 29, 2025. We hope this gives more authors the chance to contribute! More info: https://t.co/zQoPvinxHu Please share! #MLSP2025 #MachineLearning #SignalProcessing #DeepLearning #CallForPapers
0
4
9
π’ Join our Conversational AI Reading Group! π
Thursday, April 10th | 11 AM - 12 PM EST πSpeaker: Karen Livescu from TTIC π Topic: "Toward Understanding Sign Language in the Real World" π Details: ( https://t.co/oxEZtl9zYt)
0
2
3
Huge thanks to @NVIDIAAIDev for the NVIDIA Academic Grant #NVIDIAGrant! My team is already using the donated GPU hours on Saturn Cloud to accelerate our research. So grateful! #research #AI #SpeechBrain #GPU #A100 #DeepLearning #OpenSource #NVIDIA
0
1
18
Amazing Event!
π’ Join our Conversational AI Reading Group! π
Thursday, March 13 | 11 AM - 12 PM EST πSpeaker: Alexandre Defossez @honualx π Topic: "Moshi: a speech-text foundation model for real-time dialogue" π Details: ( https://t.co/oxEZtl9zYt)
0
0
2
π’ Join our Conversational AI Reading Group! π
Thursday, March 13 | 11 AM - 12 PM EST πSpeaker: Alexandre Defossez @honualx π Topic: "Moshi: a speech-text foundation model for real-time dialogue" π Details: ( https://t.co/oxEZtl9zYt)
0
3
13
#IEEE MLSP 2025 β Call for Papers The OpenReview portal is now open! π
Deadline: April 22, 2025 π Website: https://t.co/aQm0lJGfvW We welcome #papers on #deeplearning, #speech/#audio, #LLMs, #CV, #NLP, #EEG, #optimization, #fintech, #healthcare, #wireless comms, & more!
0
3
10
As an organizer of #IEEE #MLSP2025, I can say weβre preparing an amazing event in a wonderful city (#Δ°stanbul ) We expect many speech papers as well. If you missed #Interspeech, submit to MLSP! Deadline: April 22 β Donβt miss it! https://t.co/S0Xvsm8O58
1
3
14
My lab is running an amazing Reading Group on #ConversationalAI, open to all! This week (Feb 6, 11 AM EST) we host Hung-yi Lee, a top #AI researcher! Heβll talk about: "Teaching Foundation Models New Skills: Insights and Experiences". π Website: https://t.co/jVmfl49X86
0
4
17
π’ Join us for the Conversational AI Reading Group meeting on Thursday, February 6th, from 11 AM-12 PM EST. @HungyiLee2 will present "Teaching Foundation Models New Skills: Insights and Experiences". Details here:
0
4
6
Our Speech-LLM diagnosis paper "What Are They Doing? Joint Audio-Speech Co-Reasoning" is accepted at #ICASSP 2025. Your model behaves more like an Audio-LLM or a Speech-LLM? See our specially designed task and dataset: Paper: https://t.co/aihhVxLzqN Github: https://t.co/qLefV92O7i
0
4
17