ASR

Technology that converts spoken language into written text. Automatic Speech Recognition (ASR) is an AI system that processes audio signals containing speech and transcribes them into text. ASR uses a pipeline of acoustic modelling (what sounds were produced), language modelling (what words are most likely), and decoding to produce accurate transcriptions.

Modern ASR systems use deep learning — particularly transformer-based models like OpenAI Whisper — and achieve human-level accuracy on clean audio for major languages.

Automatic Speech Recognition is a foundational concept in Application. Technology that converts spoken language into written text.