Amazon Transcribe reportedly has 20 to 50 percent accuracy across many languages and offers features such as automatic punctuation, custom vocabulary, automatic language identification, and custom vocabulary filters. It can recognize speech in audio and video formats and in noisy environments. The improvements in Amazon Transcribe also enhance the accuracy of its Call Analytics platform, often used by contact center customers. The platform, now powered by generative AI models, summarizes interactions between an agent and a customer, reducing after-call report creation work and allowing managers to quickly access information without needing to read the entire transcript.
Key takeaways:
- AWS has added new languages to its Amazon Transcribe product, now offering AI-based transcription for 100 languages.
- Amazon Transcribe has been trained on millions of hours of unlabeled audio data from over 100 languages, using self-supervised algorithms to learn patterns of human speech in different languages and accents.
- The transcription service offers 20 to 50 percent accuracy across many languages, and includes features like automatic punctuation, custom vocabulary, automatic language identification, and custom vocabulary filters.
- Improvements in Amazon Transcribe also enhance the accuracy of its Call Analytics platform, reducing after-call work for creating reports and allowing managers to quickly access information without needing to read the entire transcript.