Speech, language, and semantics

Speech, Language, and Semantics

Speech vs. Language as terms of art in the ML community (also "multimodal") Language as symbol sequences Words, lexemes, characters (what's the name of the languages where symbols represent concepts vs. sound? e.g. kanjii) Accents, ligatures, diacritics Phonemes Prosody Text tokenizers - https://platform.openai.com/tokenizer