Data Annotation

Our Data Annotation Services

As the world scrambles to build accurate machine-learning models that meet our ever-growing digital needs, the lack of high-quality annotated data has become increasingly apparent.  

Data annotation is the process by which a set of data is labelled with tags that are relevant to it to train machine-learning models. Such data can take the form of audio, images, video or text, all of which must be labelled as accurately as possible.

CaptionCube has annotated more than 830 hours of audio data in Singaporean English, Mandarin Chinese, Chinese dialects and Bahasa Melayu (Malay), which includes data for the medical sector. We also have substantial experience in annotating image and video data.  

Based in Singapore, CaptionCube is home to a team of meticulous talents who are familiar with the Singaporean context and well-versed in English (including Singlish!), Mandarin Chinese and Bahasa Melayu (Malay), making us the ideal solution to your local annotation needs.  

Lighten your data engineers’ massive workloads by outsourcing the tedious task of poring over huge amounts of data to us! Let us assist you in building accurate machine-learning models in a swift and fuss-free manner.

Readily available Singaporean Speech Data Set

We have a set of full-verbatim Singaporean Conversational English speech data (100 recorded hours), which is available for licensing.