Amazon Web Services (AWS) has expanded the speech recognition capabilities of its transcription tool. Amazon Transcribe, an automatic speech recognition (ASR) service, now supports more than 100 languages thanks to a new speech-based model. AWS services leverage machine learning technology to automatically convert speech to text for businesses. Amazon Transcribe is used to transcribe customer calls, create captions and meeting notes, and detect harmful content in audio.
Announced at Sunday’s AWS re:Invent event, Amazon Transcribe’s new model was trained using millions of hours of unlabeled audio data from more than 100 languages. At the event, the company showcased a number of its AI services, now enhanced with FM (Fundamental Model)-based capabilities. . According to AWS, its transcription tools can now learn the unique patterns of human speech across a variety of languages and accents. The tool now ensures that even traditionally underrepresented languages are recognized and transcribed accurately.
According to AWS, the new model improves the transcription service’s accuracy by 20 to 50 percent for most languages. “In addition to significant accuracy improvements, this large-scale ASR model also improves readability with more accurate punctuation and capitalization,” the AWS blog announcing the new speech model said.
The new ASR model provides improved usability, customization, user safety, and privacy across all 100+ languages, AWS said. “This includes features such as automatic punctuation, custom vocabulary, automatic language identification, speaker segmentation, word-level confidence scores, and custom vocabulary filters,” he added. AWS customers can now embed speech-to-text tools into their applications that provide expanded support for different accents, noise environments, and acoustic conditions.
AWS customers using the Amazon Transcribe service in batch mode can access FM-based speech recognition without changing API endpoints or input parameters, the blog said.
In addition to improvements to Amazon Transcribe, AWS has also made improvements to other AI services. Amazon Personalize, which provides machine learning-based personalization to your customers, can now create more engaging content and product recommendations. And the Amazon Lex AI chatbot now provides accurate, conversational responses to FAQs.
Last June, there were reports that AWS was considering using new AI chips from Advanced Micro Devices (AMD). “We’re still collaborating between AWS and AMD on exactly where that will be located, but that’s something our teams are collaborating on,” said Dave Brown, Amazon’s vice president of elastic computing cloud at the time.
Affiliate links may be generated automatically. Please see our Ethics Statement for more information.
