AWS expands language capabilities in Amazon Transcribe
- AWS has announced that it is adding new languages to its Amazon Transcribe product.
- Amazon Transcribe now offers generative AI-based transcription for over 100 languages.
- The new AI capabilities in Transcribe can generate a call transcription for customer use.
- Customers can use Amazon Transcribe to add speech-to-text capabilities to their apps on the AWS Cloud.
- Transcribe has been trained on millions of hours of unlabeled audio data from over 100 languages.
- AWS ensured that some languages were not over-represented in the training data.
Introduction
AWS has expanded the language capabilities of its Amazon Transcribe product, allowing for generative AI-based transcription in over 100 languages. The new capabilities were announced during the AWS re:Invent event, and are aimed at offering customers more options for adding speech-to-text capabilities to their applications on the AWS Cloud.
Recognizing More Languages
The updated version of Amazon Transcribe is now able to recognize and transcribe speech in over 100 languages. This expansion in language recognition will provide users with a wider range of options for accurately transcribing audio content in different languages and accents. The expanded language capabilities were achieved through the use of self-supervised algorithms and training on millions of hours of unlabeled audio data from various languages.
The training data used by Amazon Transcribe ensures that not only the most commonly spoken languages are recognized accurately, but also languages that are less commonly spoken. This is important to provide accurate transcriptions and to cater to users who might need transcription services in less common languages.
New AI Capabilities
In addition to the expanded language recognition, Amazon Transcribe now offers new AI capabilities that allow for call transcription. This means that users can now transcribe phone calls, providing a valuable tool for businesses and organizations that need to keep records of important conversations.
The AI capabilities in Amazon Transcribe are based on generative AI, which means that the system is able to generate speech based on the patterns it has learned from the training data. This allows for more accurate transcriptions and better handling of accents.
