๐ง๐ต๐ฒ ๐ฎ๐ป๐ฑ ๐ ๐๐น๐๐ถ๐น๐ถ๐ป๐ด๐๐ฎ๐น ๐๐ผ๐ป๐๐ฒ๐ฟ๐๐ฎ๐๐ถ๐ผ๐ป๐ฎ๐น ๐ฆ๐ฝ๐ฒ๐ฒ๐ฐ๐ต ๐๐ฎ๐ป๐ด๐๐ฎ๐ด๐ฒ ๐ ๐ผ๐ฑ๐ฒ๐น ๐๐ต๐ฎ๐น๐น๐ฒ๐ป๐ด๐ฒ (๐ ๐๐-๐ฆ๐๐ ๐๐ต๐ฎ๐น๐น๐ฒ๐ป๐ด๐ฒ ๐ฎ๐ฌ๐ฎ๐ฒ) ๐ถ๐ ๐ป๐ผ๐ ๐ฎ๐๐๐ฟ๐ฎ๐ฐ๐๐ถ๐ป๐ด ๐ฎ๐ฐ๐๐ถ๐๐ฒ ๐ฟ๐ฒ๐ด๐ถ๐๐๐ฟ๐ฎ๐๐ถ๐ผ๐ป!
LOS ANGELES, CA, UNITED STATES, May 18, 2026 /EINPresswire.com/ — With the rapid development of large language models (LLMs) and speech language models (Speech LLMs), speech recognition and spoken language understanding are moving toward unified modeling. However, real-world multilingual conversational scenarios still present major challenges, including ๐น๐ฎ๐ป๐ด๐๐ฎ๐ด๐ฒ ๐ฑ๐ถ๐๐ฒ๐ฟ๐๐ถ๐๐, ๐ฎ๐ฐ๐ฐ๐ฒ๐ป๐ ๐๐ฎ๐ฟ๐ถ๐ฎ๐๐ถ๐ผ๐ป, ๐๐ฝ๐ฒ๐ฎ๐ธ๐ฒ๐ฟ ๐๐๐ฟ๐ป๐, ๐ฐ๐ผ๐บ๐ฝ๐น๐ฒ๐ ๐ฑ๐ถ๐ฎ๐น๐ผ๐ด๐๐ฒ ๐๐๐ฟ๐๐ฐ๐๐๐ฟ๐ฒ๐, ๐ฎ๐ป๐ฑ ๐ถ๐ป๐๐๐ณ๐ณ๐ถ๐ฐ๐ถ๐ฒ๐ป๐ ๐๐ฒ๐บ๐ฎ๐ป๐๐ถ๐ฐ ๐๐ป๐ฑ๐ฒ๐ฟ๐๐๐ฎ๐ป๐ฑ๐ถ๐ป๐ด. Results from the first MLC-SLM Challenge showed that Speech LLMs have achieved strong performance in speech recognition, while there remains significant room for further exploration in speaker diarization and deeper speech understanding for complex multilingual conversations. Building on this, the 2nd MLC-SLM Challenge aims to further advance Speech LLMs in ๐๐ฝ๐ฒ๐ฎ๐ธ๐ฒ๐ฟ ๐ฑ๐ถ๐ฎ๐ฟ๐ถ๐๐ฎ๐๐ถ๐ผ๐ป, ๐ฎ๐ฐ๐ผ๐๐๐๐ถ๐ฐ ๐๐ป๐ฑ๐ฒ๐ฟ๐๐๐ฎ๐ป๐ฑ๐ถ๐ป๐ด, ๐ฎ๐ป๐ฑ ๐๐ฒ๐บ๐ฎ๐ป๐๐ถ๐ฐ ๐๐ป๐ฑ๐ฒ๐ฟ๐๐๐ฎ๐ป๐ฑ๐ถ๐ป๐ด.
The training set for this yearโs challenge has been further expanded from the first edition, adding more language variants and accents such as ๐๐ฎ๐ป๐ฎ๐ฑ๐ถ๐ฎ๐ป ๐๐ฟ๐ฒ๐ป๐ฐ๐ต, ๐ ๐ฒ๐ ๐ถ๐ฐ๐ฎ๐ป ๐ฆ๐ฝ๐ฎ๐ป๐ถ๐๐ต, ๐ฎ๐ป๐ฑ ๐๐ฟ๐ฎ๐๐ถ๐น๐ถ๐ฎ๐ป ๐ฃ๐ผ๐ฟ๐๐๐ด๐๐ฒ๐๐ฒ. The training data totals approximately 2,100 hours and covers around 14 languages, providing richer and more realistic data support for research on multilingual conversational speech language models.
๐ ๐ฎ๐ท๐ผ๐ฟ ๐๐ฝ๐ฑ๐ฎ๐๐ฒ: ๐๐ต๐ฒ ๐ผ๐ณ๐ณ๐ถ๐ฐ๐ถ๐ฎ๐น ๐ฏ๐ฎ๐๐ฒ๐น๐ถ๐ป๐ฒ ๐๐๐๐๐ฒ๐บ๐ ๐ณ๐ผ๐ฟ ๐๐ต๐ถ๐ ๐๐ฒ๐ฎ๐ฟโ๐ ๐ฐ๐ต๐ฎ๐น๐น๐ฒ๐ป๐ด๐ฒ ๐ต๐ฎ๐๐ฒ ๐ป๐ผ๐ ๐ฏ๐ฒ๐ฒ๐ป ๐ฟ๐ฒ๐น๐ฒ๐ฎ๐๐ฒ๐ฑ!
Task 1 focuses on multilingual ๐ฐ๐ผ๐ป๐๐ฒ๐ฟ๐๐ฎ๐๐ถ๐ผ๐ป๐ฎ๐น ๐๐ฝ๐ฒ๐ฒ๐ฐ๐ต ๐๐ฝ๐ฒ๐ฎ๐ธ๐ฒ๐ฟ ๐ฑ๐ถ๐ฎ๐ฟ๐ถ๐๐ฎ๐๐ถ๐ผ๐ป ๐ฎ๐ป๐ฑ ๐ฟ๐ฒ๐ฐ๐ผ๐ด๐ป๐ถ๐๐ถ๐ผ๐ป. The baseline system is built on ๐ ๐ถ๐ฐ๐ฟ๐ผ๐๐ผ๐ณ๐โ๐ ๐ผ๐ฝ๐ฒ๐ป-๐๐ผ๐๐ฟ๐ฐ๐ฒ ๐ฉ๐ถ๐ฏ๐ฒ๐ฉ๐ผ๐ถ๐ฐ๐ฒ-๐๐ฆ๐ฅ ๐บ๐ผ๐ฑ๐ฒ๐น and fine-tuned with the challenge training set.
Task 2 focuses on ๐บ๐๐น๐๐ถ๐น๐ถ๐ป๐ด๐๐ฎ๐น ๐ฐ๐ผ๐ป๐๐ฒ๐ฟ๐๐ฎ๐๐ถ๐ผ๐ป๐ฎ๐น ๐๐ฝ๐ฒ๐ฒ๐ฐ๐ต ๐๐ป๐ฑ๐ฒ๐ฟ๐๐๐ฎ๐ป๐ฑ๐ถ๐ป๐ด. The baseline system uses ๐๐ฒ๐บ๐ถ๐ป๐ถ ๐ฎ.๐ฑ ๐ฃ๐ฟ๐ผ to construct multiple-choice questions for acoustic and semantic understanding, and is fine-tuned based on ๐ค๐๐ฒ๐ป๐ฎ.๐ฑ-๐ข๐บ๐ป๐ถ-๐ณ๐ ๐ฎ๐ป๐ฑ ๐๐ต๐ฒ ๐บ๐-๐๐๐ถ๐ณ๐ ๐๐ผ๐ผ๐น๐ธ๐ถ๐.
Participating teams can now refer to the official baseline systems to accelerate ๐๐๐๐๐ฒ๐บ ๐ฑ๐ฒ๐๐ฒ๐น๐ผ๐ฝ๐บ๐ฒ๐ป๐, ๐ฒ๐ ๐ฝ๐ฒ๐ฟ๐ถ๐บ๐ฒ๐ป๐๐ฎ๐น ๐๐ฎ๐น๐ถ๐ฑ๐ฎ๐๐ถ๐ผ๐ป, ๐ฎ๐ป๐ฑ ๐บ๐ผ๐ฑ๐ฒ๐น ๐ผ๐ฝ๐๐ถ๐บ๐ถ๐๐ฎ๐๐ถ๐ผ๐ป.
๐ง๐ต๐ฒ ๐๐ผ๐๐ฎ๐น ๐ฝ๐ฟ๐ถ๐๐ฒ ๐ฝ๐ผ๐ผ๐น ๐ณ๐ผ๐ฟ ๐๐ต๐ถ๐ ๐๐ฒ๐ฎ๐ฟโ๐ ๐ฐ๐ต๐ฎ๐น๐น๐ฒ๐ป๐ด๐ฒ ๐ถ๐ ๐จ๐ฆ๐ ๐ฎ๐ฌ,๐ฌ๐ฌ๐ฌ ๐ผ๐ฟ ๐ฒ๐พ๐๐ถ๐๐ฎ๐น๐ฒ๐ป๐ ๐ถ๐ป ๐ผ๐๐ต๐ฒ๐ฟ ๐ฐ๐๐ฟ๐ฟ๐ฒ๐ป๐ฐ๐ถ๐ฒ๐!
Outstanding teams will receive awards for each task:
๐ฅ 1st Place: ๐จ๐ฆ๐ ๐ฑ,๐ฌ๐ฌ๐ฌ
๐ฅ 2nd Place: ๐จ๐ฆ๐ ๐ฏ,๐ฌ๐ฌ๐ฌ
๐ฅ 3rd Place: ๐จ๐ฆ๐ ๐ฎ,๐ฌ๐ฌ๐ฌ
Whether focusing on speaker diarization and recognition or multilingual conversational speech understanding, participating teams will have the opportunity to showcase their technical solutions and compete for awards on an international challenge platform.
The challenge is attracting increasing participation from both academia and industry, including teams from ๐น๐ฒ๐ฎ๐ฑ๐ถ๐ป๐ด ๐ด๐น๐ผ๐ฏ๐ฎ๐น ๐ฐ๐ผ๐บ๐ฝ๐ฎ๐ป๐ถ๐ฒ๐ ๐ฎ๐ป๐ฑ ๐ฟ๐ฒ๐ป๐ผ๐๐ป๐ฒ๐ฑ ๐๐ป๐ถ๐๐ฒ๐ฟ๐๐ถ๐๐ถ๐ฒ๐. This growing engagement reflects the strong momentum behind multilingual speech language model technologies across research and real-world applications.
Whether you work on speech recognition, speaker diarization, speech understanding, multimodal large models, or multilingual data and evaluation, MLC-SLM offers a platform to compete and collaborate with researchers, engineers, and industry teams from around the world.
We welcome ๐๐ป๐ถ๐๐ฒ๐ฟ๐๐ถ๐๐ถ๐ฒ๐, ๐ฟ๐ฒ๐๐ฒ๐ฎ๐ฟ๐ฐ๐ต ๐ถ๐ป๐๐๐ถ๐๐๐๐ถ๐ผ๐ป๐, ๐ฒ๐ป๐๐ฒ๐ฟ๐ฝ๐ฟ๐ถ๐๐ฒ ๐๐ฒ๐ฎ๐บ๐, ๐ฎ๐ป๐ฑ ๐ถ๐ป๐ฑ๐ถ๐๐ถ๐ฑ๐๐ฎ๐น ๐ฟ๐ฒ๐๐ฒ๐ฎ๐ฟ๐ฐ๐ต๐ฒ๐ฟ๐ to register and participate. Join us in advancing the development of multilingual conversational speech language models!
Registration is ongoing. We look forward to your participation.
Official Website Link: https://www.nexdata.ai/competition/mlc-slm
Registration Link: https://forms.gle/jfAZ95abGy4ZiNHo7
NEXDATA
MLC-SLM Competition Committee
mlc-slmw@nexdata.ai
Legal Disclaimer:
EIN Presswire provides this news content “as is” without warranty of any kind. We do not accept any responsibility or liability
for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this
article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
Media gallery
