Earlier this month, the cloud division of technology giant Amazon, Amazon Web Services, Inc. (AWS), launched two new language-based machine learning services, Amazon Transcribe and Amazon Translate.
According to AWS, Amazon Translate is a “deep learning-powered machine translation service that provides natural-sounding language translation in both real-time and batch scenarios.”
Amazon Transcribe helps users turn spoken-word audio files into “grammatically correct” transcriptions with text that can then be “analyzed, indexed, and searched.”
Amazon Translation and Amazon Transcribe expand upon the language functions already available through AWS offerings Amazon Lex (conversational interfaces), Amazon Polly (text-to-speech), and Amazon Comprehend (natural language processing).
In its rollout, the company also announced that “tens of thousands of customers” are now using services within its wider machine learning portfolio of offerings, an increase of 250% since last year in terms of active users.
The company credits the rapid uptake to its launch of Amazon SageMaker, which made a big splash at its AWS re: Invent 2017 event last year. The solution aims to make it easier to adopt and implement machine learning-related processes into an organization.
AWS said in a statement that in addition to SageMaker, clients are using its P2 and P3 graphical processing unit (GPU) instances, deep learning Amazon Machine Images (AMIs), and AWS DeepLens.
“A lot of companies are talking about the potential of machine learning and artificial intelligence, and thinking about how to incorporate these technologies in their applications,” said Swami Sivasubramanian, vice president of machine learning at AWS.
“But, in reality, machine learning has been out of reach for all but the few organizations who have expert practitioners and data scientists on staff. AWS changed all this with the introduction of Amazon SageMaker, which makes machine learning accessible to everyday developers by eliminating the heavy lifting of building, training, and deploying models.”
(Photo credit: olilynch / Pixabay)