Case study for autonomous driving

The Company

A global leading OEM is look forward to train its voice recognition system with speech data over 30 languages.

The Challenge

This global automotive manufacturers face the complexity of localizing their in-car voice assistant systems for multiple languages. This requires a huge amount data collections for their team to manage. However, the problem should be difficult. Firstly, they should think about the speech should collected in multiple driving environments to meet the needs about recognize different scenarios for driving, such as weather conditions, traffic conditions and so on. Another is that language background will also make the difficulties for them, such as the local language expression habits, these will require the experience of linguistic expert.

The solution

Nexdata owned a deep experience on data solution for automotive industry. Nexdata provides the highest fully managed workforce throughout the entire data pipeline. Our skilled data analysts apply best practices developed from annotating millions of speech data to deliver best-in-class data labeling for autonomous vehicle companies across the world. Our experienced project managers work directly with the OEM’s engineering team to define the program, localize commands, recruit native speakers, record in various environments... We also provide professional recording equipments to collect utterances in various driving environments. Second, our team of highly experienced linguists is on hand to provide control and advice at all stages of the project, ensuring not only data consistency, but also localization value.

The Result

After working with us, the OEM team was able to develop powerful in-vehicle system programs faster, allowing the customer to seize market opportunities faster, and a long-term relationship has been reached since.