MMTEB: Massive Multilingual Text Embedding Benchmark
Paper • 2502.13595 • Published • 50
Datasets:
id string | text string |
|---|---|
0 | This is a sound of dog |
1 | This is a sound of rooster |
2 | This is a sound of pig |
3 | This is a sound of cow |
4 | This is a sound of frog |
5 | This is a sound of cat |
6 | This is a sound of hen |
7 | This is a sound of insects |
8 | This is a sound of sheep |
9 | This is a sound of crow |
10 | This is a sound of rain |
11 | This is a sound of sea_waves |
12 | This is a sound of crackling_fire |
13 | This is a sound of crickets |
14 | This is a sound of chirping_birds |
15 | This is a sound of water_drops |
16 | This is a sound of wind |
17 | This is a sound of pouring_water |
18 | This is a sound of toilet_flush |
19 | This is a sound of thunderstorm |
20 | This is a sound of crying_baby |
21 | This is a sound of sneezing |
22 | This is a sound of clapping |
23 | This is a sound of breathing |
24 | This is a sound of coughing |
25 | This is a sound of footsteps |
26 | This is a sound of laughing |
27 | This is a sound of brushing_teeth |
28 | This is a sound of snoring |
29 | This is a sound of drinking_sipping |
30 | This is a sound of door_wood_knock |
31 | This is a sound of mouse_click |
32 | This is a sound of keyboard_typing |
33 | This is a sound of door_wood_creaks |
34 | This is a sound of can_opening |
35 | This is a sound of washing_machine |
36 | This is a sound of vacuum_cleaner |
37 | This is a sound of clock_alarm |
38 | This is a sound of clock_tick |
39 | This is a sound of glass_breaking |
40 | This is a sound of helicopter |
41 | This is a sound of chainsaw |
42 | This is a sound of siren |
43 | This is a sound of car_horn |
44 | This is a sound of engine |
45 | This is a sound of train |
46 | This is a sound of church_bells |
47 | This is a sound of airplane |
48 | This is a sound of fireworks |
49 | This is a sound of hand_saw |
Environmental Sound Classification Dataset.
| Task category | AudioZeroshotClassification (audio-to-text) |
| Domains | Spoken |
| Reference | Proceedings of the 23rd {Annual ACM Conference |
Source datasets:
How to evaluate on this task
You can evaluate an embedding model on this dataset using the following code:
import mteb
task = mteb.get_task("ESC50_Zeroshot")
model = mteb.get_model(YOUR_MODEL)
mteb.evaluate(model, task)
To learn more about how to run models on mteb task check out the GitHub repository.
Citation
If you use this dataset, please cite the dataset as well as mteb, as this dataset likely includes additional processing as a part of the MMTEB Contribution.
@inproceedings{piczak2015dataset,
author = {Piczak, Karol J.},
booktitle = {Proceedings of the 23rd {Annual ACM Conference} on {Multimedia}},
date = {2015-10-13},
doi = {10.1145/2733373.2806390},
isbn = {978-1-4503-3459-4},
location = {{Brisbane, Australia}},
pages = {1015--1018},
publisher = {{ACM Press}},
title = {{ESC}: {Dataset} for {Environmental Sound Classification}},
url = {http://dl.acm.org/citation.cfm?doid=2733373.2806390},
}
@article{enevoldsen2025mmtebmassivemultilingualtext,
title={MMTEB: Massive Multilingual Text Embedding Benchmark},
author={Kenneth Enevoldsen and Isaac Chung and Imene Kerboua and Márton Kardos and Ashwin Mathur and David Stap and Jay Gala and Wissam Siblini and Dominik Krzemiński and Genta Indra Winata and Saba Sturua and Saiteja Utpala and Mathieu Ciancone and Marion Schaeffer and Gabriel Sequeira and Diganta Misra and Shreeya Dhakal and Jonathan Rystrøm and Roman Solomatin and Ömer Çağatan and Akash Kundu and Martin Bernstorff and Shitao Xiao and Akshita Sukhlecha and Bhavish Pahwa and Rafał Poświata and Kranthi Kiran GV and Shawon Ashraf and Daniel Auras and Björn Plüster and Jan Philipp Harries and Loïc Magne and Isabelle Mohr and Mariya Hendriksen and Dawei Zhu and Hippolyte Gisserot-Boukhlef and Tom Aarsen and Jan Kostkan and Konrad Wojtasik and Taemin Lee and Marek Šuppa and Crystina Zhang and Roberta Rocca and Mohammed Hamdy and Andrianos Michail and John Yang and Manuel Faysse and Aleksei Vatolin and Nandan Thakur and Manan Dey and Dipam Vasani and Pranjal Chitale and Simone Tedeschi and Nguyen Tai and Artem Snegirev and Michael Günther and Mengzhou Xia and Weijia Shi and Xing Han Lù and Jordan Clive and Gayatri Krishnakumar and Anna Maksimova and Silvan Wehrli and Maria Tikhonova and Henil Panchal and Aleksandr Abramov and Malte Ostendorff and Zheng Liu and Simon Clematide and Lester James Miranda and Alena Fenogenova and Guangyu Song and Ruqiya Bin Safi and Wen-Ding Li and Alessia Borghini and Federico Cassano and Hongjin Su and Jimmy Lin and Howard Yen and Lasse Hansen and Sara Hooker and Chenghao Xiao and Vaibhav Adlakha and Orion Weller and Siva Reddy and Niklas Muennighoff},
publisher = {arXiv},
journal={arXiv preprint arXiv:2502.13595},
year={2025},
url={https://arxiv.org/abs/2502.13595},
doi = {10.48550/arXiv.2502.13595},
}
@article{muennighoff2022mteb,
author = {Muennighoff, Niklas and Tazi, Nouamane and Magne, Loïc and Reimers, Nils},
title = {MTEB: Massive Text Embedding Benchmark},
publisher = {arXiv},
journal={arXiv preprint arXiv:2210.07316},
year = {2022}
url = {https://arxiv.org/abs/2210.07316},
doi = {10.48550/ARXIV.2210.07316},
}
Dataset Statistics
Dataset Statistics
The following code contains the descriptive statistics from the task. These can also be obtained using:
import mteb
task = mteb.get_task("ESC50_Zeroshot")
desc_stats = task.metadata.descriptive_stats
{
"train": {
"num_samples": 2000,
"number_of_characters": null,
"text_statistics": null,
"image_statistics": null,
"audio_statistics": {
"total_duration_seconds": 10000.0,
"min_duration_seconds": 5.0,
"average_duration_seconds": 5.0,
"max_duration_seconds": 5.0,
"unique_audios": 2000,
"average_sampling_rate": 44100.0,
"sampling_rates": {
"44100": 2000
}
},
"label_statistics": {
"min_labels_per_text": 1,
"average_label_per_text": 1.0,
"max_labels_per_text": 1,
"unique_labels": 50,
"labels": {
"0": {
"count": 40
},
"14": {
"count": 40
},
"36": {
"count": 40
},
"19": {
"count": 40
},
"30": {
"count": 40
},
"34": {
"count": 40
},
"9": {
"count": 40
},
"22": {
"count": 40
},
"48": {
"count": 40
},
"41": {
"count": 40
},
"47": {
"count": 40
},
"31": {
"count": 40
},
"17": {
"count": 40
},
"45": {
"count": 40
},
"8": {
"count": 40
},
"15": {
"count": 40
},
"46": {
"count": 40
},
"37": {
"count": 40
},
"32": {
"count": 40
},
"16": {
"count": 40
},
"25": {
"count": 40
},
"4": {
"count": 40
},
"3": {
"count": 40
},
"27": {
"count": 40
},
"43": {
"count": 40
},
"12": {
"count": 40
},
"40": {
"count": 40
},
"29": {
"count": 40
},
"10": {
"count": 40
},
"7": {
"count": 40
},
"26": {
"count": 40
},
"6": {
"count": 40
},
"44": {
"count": 40
},
"23": {
"count": 40
},
"20": {
"count": 40
},
"49": {
"count": 40
},
"24": {
"count": 40
},
"39": {
"count": 40
},
"28": {
"count": 40
},
"18": {
"count": 40
},
"2": {
"count": 40
},
"35": {
"count": 40
},
"38": {
"count": 40
},
"21": {
"count": 40
},
"1": {
"count": 40
},
"11": {
"count": 40
},
"42": {
"count": 40
},
"5": {
"count": 40
},
"33": {
"count": 40
},
"13": {
"count": 40
}
}
},
"candidates_labels_text_statistics": {
"total_text_length": 1402,
"min_text_length": 22,
"average_text_length": 28.04,
"max_text_length": 35,
"unique_texts": 50
}
}
}
This dataset card was automatically generated using MTEB
- Downloads last month
- 50