How HMM speech recognizers modeled variable timing, learned from recordings, and combined acoustic evidence with dictionaries and language models.
How Multilingual Speech Recognizers Worked Before Deep Learning
A look at multilingual speech recognition in the 1990s and 2000s: shared acoustic models, lexicons, language identification, code-switching and evaluation.
How Conferences Shaped Pattern-Recognition Research in the 2000s
How 2000s conferences shaped pattern-recognition research through shared benchmarks, short papers, workshops, demonstrations, and peer scrutiny.
Language Processing Research in the 2000s: Data, Benchmarks, and Pipelines
How 2000s language-processing researchers used statistical models, shared tasks, web text, and multilingual data—and why annotation and evaluation still mattered.
Biometric Recognition in the Mid-2000s: Capture, Matching and Evaluation
How mid-2000s researchers evaluated face, iris, hand, voice and gait recognition—and why capture failures mattered as much as match scores.
How Early Speech Researchers Built Voices from Acoustic Patterns
A look at pattern playback, formant synthesis, coarticulation, prosody, and the listening tests that shaped early artificial speech.
How Early Object Trackers Kept Sight of Identity
A look at late-1990s and early-2000s object tracking: motion models, appearance cues, occlusion, camera geometry, and the difficulty of measuring identity.
Early Low-Resource Speech Recognition: Building Data, Methods, and Evidence
How early academic speech-recognition projects worked with under-resourced languages through corpus building, acoustic transfer, careful transcription, and credible evaluation.
Speaker Diarization in the 2000s: Solving the “Who Spoke When?” Problem
How 2000s speaker diarization systems used speech detection, BIC segmentation, clustering, GMMs, and HMM resegmentation to determine who spoke when.
Reading 2000s Pattern Recognition Conference Proceedings
How mid-2000s pattern recognition conferences documented speech, vision, biometrics, medical imaging, datasets, evaluation methods, and local research practice.