Nexdata to showcase AI training data at INTERSPEECH 2026

Nexdata will exhibit at INTERSPEECH 2026 in Sydney from Sept. 27 to Oct. 1 and highlight multilingual speech, Speech LLM, LLM and multimodal AI training data at Booth E5. The company will also point attendees to the Oct. 2 MLC-SLM Workshop, where researchers will discuss multilingual conversational speech and Speech LLM development.

Why it matters: - Nexdata is targeting builders of speech and multimodal AI at a major industry conference where data quality and language coverage can shape model performance. - The company is emphasizing multilingual and multi-accent data as demand grows for more natural human-AI interaction.

What happened: - Nexdata will exhibit at INTERSPEECH 2026 from Sept. 27 to Oct. 1 at ICC Sydney in Sydney, Australia. - Nexdata will be at Booth #E5. - The company will showcase data resources and services for Speech LLMs, LLMs and multimodal AI. - Nexdata will also highlight multilingual speech understanding and natural human-AI interaction use cases.

The details: - Featured data capabilities include natural conversations, multilingual and multi-accent speech, multi-speaker conversations, long-form speech, Speech LLM training data and multimodal data. - The data is positioned to support speech recognition, speech understanding, conversational AI and multimodal model training. - Nexdata covers 200+ languages and dialects. - Nexdata offers off-the-shelf datasets and bespoke data services. - Bespoke projects can be tailored by language, accent, scenario, task, data format, collection and annotation. - The 2nd Multilingual Conversational Speech Language Model Workshop will follow in Sydney on Oct. 2, 2026. - The workshop will bring together researchers and industry experts focused on multilingual conversational speech and Speech LLM technologies. - The full-day workshop will include five keynote presentations, the 2nd MLC-SLM Challenge 2026 summary and awards ceremony, oral sessions and a poster session. - Keynote speakers include Yonghong Yan of the Institute of Acoustics, Chinese Academy of Sciences; Jinyu Li of Microsoft; Kai Yu of Shanghai Jiao Tong University; Hung-yi Lee of National Taiwan University; and Shinnosuke Takamichi of Keio University. - The workshop will announce the final results of the 2nd MLC-SLM Challenge 2026 and recognize winning teams. - Participating teams will present systems, technical approaches and research results in multilingual conversational speech recognition and understanding. - This year’s challenge expanded language and accent coverage to Tagalog, Urdu, Turkish, Canadian French, Mexican Spanish and Brazilian Portuguese. - The competition focused on multilingual conversational speech logging and recognition, plus multilingual conversational speech understanding. - More information about the 2nd MLC-SLM Challenge and Workshop is available on the Nexdata website. - Nexdata was founded in 2011. - Nexdata says its services also support Generative AI, Conversational AI, Physical AI and autonomous driving. - Nexdata’s social links include LinkedIn, Facebook, X and YouTube.

Between the lines: - Nexdata is pairing its conference presence with a related workshop to reinforce its role in the multilingual speech ecosystem. - The expanded language list in the MLC-SLM Challenge shows continued focus on broader geographic and accent coverage.

What's next: - INTERSPEECH attendees can visit Booth #E5 in Sydney to discuss data requirements for Speech LLMs, multilingual speech, LLMs and multimodal AI. - The 2nd MLC-SLM Workshop will take place on Oct. 2, 2026, in Sydney. - The challenge’s final results and winning teams will be announced at the workshop.