Automatic Speech Recognition (ASR) converts spoken language into text through neural models that process acoustic features, align temporal sequences, and decode linguistic content. ASR powers voice assistants, transcription services, accessibility tools, and real-time communication platforms across 100+ languages. Modern ASR achieved a breakthrough in 2022–2025 with transformer-based architectures (Whisper, Conformer) and self-supervised pre-training (wav2vec2, HuBERT) reaching near-human accuracy on clean speech, though challenges remain in noisy environments, accented speech, and low-resource languages. The field is split between offline models optimizing for accuracy on pre-recorded audio and streaming models balancing latency with real-time transcription—a trade-off that fundamentally shapes architecture choices from CTC-based systems to RNN-Transducers.