AWS’s transcription platform is now powered by generative AI
AWS has added new languages to its Amazon Transcribe product, providing creative AI-driven transcription for 100 languages and a host of new AI capabilities for customers.
Announced during the AWS re:Invent event, Amazon Transcribe can now recognize more spoken languages and turn on call forwarding. Customers use AWS Transcribe to add speech-to-text capabilities to their applications on the AWS cloud.
Transcribe is trained on “millions of hours of unlabeled audio data from more than 100 languages” and uses self-supervised algorithms to learn human speech patterns in different languages and dialects, the company said in a blog post. AWS said it has ensured that some languages are not overrepresented in the training data to ensure that less frequently used languages are as accurate as more frequently used languages.
As of late 2022, Amazon Transcribe supports 79 languages.
Amazon Transcribe has 20% to 50% accuracy across many languages, according to AWS. It also offers automatic punctuation, custom vocabulary, automatic language recognition, and custom vocabulary filters. It can recognize speech in audio and video formats and noisy environments.
the edge I reached out to AWS for information about the previous resolutions and underlying models I used in Amazon Transcribe.
With better language recognition, AWS said advances in Amazon Transcribe also lead to better accuracy through its Call Analytics system, which its contact center customers often use. Amazon Transcribe Call Analytics, now also powered by generative AI models, summarizes agent-customer interactions. This reduces post-call work to create reports, and managers can quickly read the information without having to review the entire text, AWS said.
Of course, AWS isn’t the only company offering AI-powered transcription services. Otter has been providing AI transcripts to consumers and businesses for a while, and released a summarization tool in June. Although that’s not quite the case, Meta has announced that it is working on an AI-powered generative translation model that recognizes nearly 100 spoken languages.
AWS also announced additional capabilities for its Amazon Personalization product, which allows customers to offer products or display recommendations to customers, such as how streaming services can suggest new shows based on past activity. AWS has added Content Creation, which will write headlines or email subject lines to topically link to recommendation lists.