CHAT (Codes for the Human Analysis of Transcripts) is the most common format for transcribing, encoding and analysing child language data. It was developed by Brian MacWhinney within the CHILDES (Child Language Data Exchange System) Project. All of the transcripts in the CHILDES database are encoded in the CHAT format.

By using the CHAT format, transcribed data can be annotated with different levels of linguistic information such as phonological, morphological or syntactic information. The data transcribed and annotated in CHAT, are compatible with some analysis tools, for example CLAN provided also by the CHILDES Project.

Since the CHAT format is widely recognized in children’s language research, many transcription tools (such as ELAN, EXMARaLDA, Praat, Phon, Transcriber) supports it. These tools can also export and import data encoded in various formats as CHAT formatted data.