Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

WavLink: Compact Audio-Text Embeddings with a Global Whisper Token (2026)
MUKA: Multi Kernel Audio Adaptation Of Audio-Language Models (2026)
Eureka-Audio: Triggering Audio Intelligence in Compact Language Models (2026)
UniAudio 2.0: A Unified Audio Language Model with Text-Aligned Factorized Audio Tokenization (2026)
LongSpeech: A Scalable Benchmark for Transcription, Translation and Understanding in Long Speech (2026)
Audio-to-Image Bird Species Retrieval without Audio-Image Pairs via Text Distillation (2026)
LEMON: How Well Do MLLMs Perform Temporal Multimodal Understanding on Instructional Videos? (2026)

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on Hugging Face checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend

\n","updatedAt":"2026-02-20T01:39:06.262Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":318,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.6944173574447632},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2602.16008","authors":[{"_id":"6996e55577d9294348451e19","user":{"_id":"6671be9ff022d14aa10df864","avatarUrl":"/avatars/dd085abefa38c1604dc2ceabf472816d.svg","isPro":false,"fullname":"Adnan El Assadi","user":"AdnanElAssadi","type":"user"},"name":"Adnan El Assadi","status":"claimed_verified","statusLastChangedAt":"2026-02-19T14:42:12.758Z","hidden":false},{"_id":"6996e55577d9294348451e1a","user":{"_id":"64cc0e80a257a3212c0c4b24","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64cc0e80a257a3212c0c4b24/wqs6WZN8-3OQthcnQXgN7.png","isPro":false,"fullname":"Isaac Chung","user":"isaacchung","type":"user"},"name":"Isaac Chung","status":"claimed_verified","statusLastChangedAt":"2026-02-19T11:04:09.879Z","hidden":false},{"_id":"6996e55577d9294348451e1b","name":"Chenghao Xiao","hidden":false},{"_id":"6996e55577d9294348451e1c","user":{"_id":"61af4544d691b3aadd1f62b6","avatarUrl":"/avatars/7a4067accdd1005f78c3c4adad3ee0a5.svg","isPro":false,"fullname":"Solomatin Roman","user":"Samoed","type":"user"},"name":"Roman Solomatin","status":"claimed_verified","statusLastChangedAt":"2026-02-19T11:04:05.803Z","hidden":false},{"_id":"6996e55577d9294348451e1d","user":{"_id":"67aaf15be193f6344f7e4390","avatarUrl":"/avatars/ec7e7a2d56fe5943b30378062a1fc4fd.svg","isPro":false,"fullname":"Animesh Jha","user":"anime-sh","type":"user"},"name":"Animesh Jha","status":"claimed_verified","statusLastChangedAt":"2026-02-20T08:37:14.884Z","hidden":false},{"_id":"6996e55577d9294348451e1e","name":"Rahul Chand","hidden":false},{"_id":"6996e55577d9294348451e1f","name":"Silky Singh","hidden":false},{"_id":"6996e55577d9294348451e20","user":{"_id":"67d0ea5d94e08c03254dfb56","avatarUrl":"/avatars/8e511f5bcebfa869e00bc644e9957162.svg","isPro":false,"fullname":"kaitlyn wang","user":"wangusbeef","type":"user"},"name":"Kaitlyn Wang","status":"claimed_verified","statusLastChangedAt":"2026-02-19T14:41:41.369Z","hidden":false},{"_id":"6996e55577d9294348451e21","user":{"_id":"66fd9e8b001816b29e0bc2e4","avatarUrl":"/avatars/190e6711cef3126c51e15276bb5ed8cd.svg","isPro":false,"fullname":"Ali Sartaz Khan","user":"alisartazkhan","type":"user"},"name":"Ali Sartaz Khan","status":"claimed_verified","statusLastChangedAt":"2026-02-20T08:37:19.294Z","hidden":false},{"_id":"6996e55577d9294348451e22","name":"Marc Moussa Nasser","hidden":false},{"_id":"6996e55577d9294348451e23","name":"Sufen Fong","hidden":false},{"_id":"6996e55577d9294348451e24","name":"Pengfei He","hidden":false},{"_id":"6996e55577d9294348451e25","name":"Alan Xiao","hidden":false},{"_id":"6996e55577d9294348451e26","user":{"_id":"6754994f0a4a1144aec6ef57","avatarUrl":"/avatars/9dc00280582bcb0ace57cb34d25e91a0.svg","isPro":false,"fullname":"Ayush Sunil Munot","user":"AyushM6","type":"user"},"name":"Ayush Sunil Munot","status":"claimed_verified","statusLastChangedAt":"2026-02-19T11:04:07.782Z","hidden":false},{"_id":"6996e55577d9294348451e27","name":"Aditya Shrivastava","hidden":false},{"_id":"6996e55577d9294348451e28","user":{"_id":"6744d6ee334ae6264b8b3d99","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/CC8d048suNVsmlA98Opmn.png","isPro":false,"fullname":"Artem G","user":"arteemg","type":"user"},"name":"Artem Gazizov","status":"claimed_verified","statusLastChangedAt":"2026-02-20T08:37:21.674Z","hidden":false},{"_id":"6996e55577d9294348451e29","name":"Niklas Muennighoff","hidden":false},{"_id":"6996e55577d9294348451e2a","user":{"_id":"5ff5943752c26e9bc240bada","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/Exyzf3C_gJ2KdsL4K5_cq.png","isPro":false,"fullname":"Kenneth C. Enevoldsen","user":"KennethEnevoldsen","type":"user"},"name":"Kenneth Enevoldsen","status":"claimed_verified","statusLastChangedAt":"2026-02-20T08:37:17.036Z","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/61af4544d691b3aadd1f62b6/vkcutGZImTRS5bB6Bk_uY.png"],"publishedAt":"2026-02-17T21:00:51.000Z","submittedOnDailyAt":"2026-02-19T07:59:30.545Z","title":"MAEB: Massive Audio Embedding Benchmark","submittedOnDailyBy":{"_id":"61af4544d691b3aadd1f62b6","avatarUrl":"/avatars/7a4067accdd1005f78c3c4adad3ee0a5.svg","isPro":false,"fullname":"Solomatin Roman","user":"Samoed","type":"user"},"summary":"We introduce the Massive Audio Embedding Benchmark (MAEB), a large-scale benchmark covering 30 tasks across speech, music, environmental sounds, and cross-modal audio-text reasoning in 100+ languages. We evaluate 50+ models and find that no single model dominates across all tasks: contrastive audio-text models excel at environmental sound classification (e.g., ESC50) but score near random on multilingual speech tasks (e.g., SIB-FLEURS), while speech-pretrained models show the opposite pattern. Clustering remains challenging for all models, with even the best-performing model achieving only modest results. We observe that models excelling on acoustic understanding often perform poorly on linguistic tasks, and vice versa. We also show that the performance of audio encoders on MAEB correlates highly with their performance when used in audio large language models. MAEB is derived from MAEB+, a collection of 98 tasks. MAEB is designed to maintain task diversity while reducing evaluation cost, and it integrates into the MTEB ecosystem for unified evaluation across text, image, and audio modalities. We release MAEB and all 98 tasks along with code and a leaderboard at https://github.com/embeddings-benchmark/mteb.","upvotes":16,"discussionId":"6996e55577d9294348451e2b","projectPage":"http://mteb-leaderboard.hf.space/?benchmark_name=MAEB%28beta%29","ai_summary":"MAEB is a large-scale audio benchmark evaluating 50+ models across 30 tasks in speech, music, and environmental sounds, revealing diverse model strengths and establishing correlations with audio LLM performance.","ai_keywords":["audio embedding benchmark","audio-text reasoning","multilingual speech tasks","acoustic understanding","linguistic tasks","audio encoders","audio large language models","MTEB ecosystem"],"organization":{"_id":"624bfda5459c48438cc39f80","name":"mteb","fullname":"Massive Text Embedding Benchmark","avatar":"https://cdn-uploads.huggingface.co/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"61af4544d691b3aadd1f62b6","avatarUrl":"/avatars/7a4067accdd1005f78c3c4adad3ee0a5.svg","isPro":false,"fullname":"Solomatin Roman","user":"Samoed","type":"user"},{"_id":"6754994f0a4a1144aec6ef57","avatarUrl":"/avatars/9dc00280582bcb0ace57cb34d25e91a0.svg","isPro":false,"fullname":"Ayush Sunil Munot","user":"AyushM6","type":"user"},{"_id":"64cc0e80a257a3212c0c4b24","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64cc0e80a257a3212c0c4b24/wqs6WZN8-3OQthcnQXgN7.png","isPro":false,"fullname":"Isaac Chung","user":"isaacchung","type":"user"},{"_id":"6671be9ff022d14aa10df864","avatarUrl":"/avatars/dd085abefa38c1604dc2ceabf472816d.svg","isPro":false,"fullname":"Adnan El Assadi","user":"AdnanElAssadi","type":"user"},{"_id":"67d0ea5d94e08c03254dfb56","avatarUrl":"/avatars/8e511f5bcebfa869e00bc644e9957162.svg","isPro":false,"fullname":"kaitlyn wang","user":"wangusbeef","type":"user"},{"_id":"5ff5943752c26e9bc240bada","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/Exyzf3C_gJ2KdsL4K5_cq.png","isPro":false,"fullname":"Kenneth C. Enevoldsen","user":"KennethEnevoldsen","type":"user"},{"_id":"63108cc834c7d77420b0fd68","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63108cc834c7d77420b0fd68/taDnqEmcI9Rhe3uzcPEE3.jpeg","isPro":false,"fullname":"Chenghao Xiao","user":"gowitheflow","type":"user"},{"_id":"5f1eb362eec0ad2a071ad6e2","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f1eb362eec0ad2a071ad6e2/IXMYkYKuTwn6kBdWnQeeY.png","isPro":false,"fullname":"Niklas Muennighoff","user":"Muennighoff","type":"user"},{"_id":"6744d6ee334ae6264b8b3d99","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/CC8d048suNVsmlA98Opmn.png","isPro":false,"fullname":"Artem G","user":"arteemg","type":"user"},{"_id":"63213080d2d45f3151837eba","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63213080d2d45f3151837eba/aBhKfY-0gZhKGQmb_Gwi2.png","isPro":true,"fullname":"Dan Jacobellis","user":"danjacobellis","type":"user"},{"_id":"609bbe2f4932693ca2009d6a","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1620819560688-609bbe2f4932693ca2009d6a.jpeg","isPro":false,"fullname":"Antoine Chaffin","user":"NohTow","type":"user"},{"_id":"66fd9e8b001816b29e0bc2e4","avatarUrl":"/avatars/190e6711cef3126c51e15276bb5ed8cd.svg","isPro":false,"fullname":"Ali Sartaz Khan","user":"alisartazkhan","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0,"organization":{"_id":"624bfda5459c48438cc39f80","name":"mteb","fullname":"Massive Text Embedding Benchmark","avatar":"https://cdn-uploads.huggingface.co/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png"}}">

Papers

arxiv:2602.16008

MAEB: Massive Audio Embedding Benchmark

Published on Feb 17

· Submitted by

Solomatin Roman on Feb 19

Massive Text Embedding Benchmark

Upvote

Authors:

Adnan El Assadi ,

Isaac Chung ,

Roman Solomatin ,

Animesh Jha ,

Kaitlyn Wang ,

Ali Sartaz Khan ,

Ayush Sunil Munot ,

Artem Gazizov ,

Kenneth Enevoldsen

Abstract

MAEB is a large-scale audio benchmark evaluating 50+ models across 30 tasks in speech, music, and environmental sounds, revealing diverse model strengths and establishing correlations with audio LLM performance.

AI-generated summary

We introduce the Massive Audio Embedding Benchmark (MAEB), a large-scale benchmark covering 30 tasks across speech, music, environmental sounds, and cross-modal audio-text reasoning in 100+ languages. We evaluate 50+ models and find that no single model dominates across all tasks: contrastive audio-text models excel at environmental sound classification (e.g., ESC50) but score near random on multilingual speech tasks (e.g., SIB-FLEURS), while speech-pretrained models show the opposite pattern. Clustering remains challenging for all models, with even the best-performing model achieving only modest results. We observe that models excelling on acoustic understanding often perform poorly on linguistic tasks, and vice versa. We also show that the performance of audio encoders on MAEB correlates highly with their performance when used in audio large language models. MAEB is derived from MAEB+, a collection of 98 tasks. MAEB is designed to maintain task diversity while reducing evaluation cost, and it integrates into the MTEB ecosystem for unified evaluation across text, image, and audio modalities. We release MAEB and all 98 tasks along with code and a leaderboard at https://github.com/embeddings-benchmark/mteb.