Synthesis and recognition demand language knowledge databases that coding does not.

Evidence

Speech synthesis and recognition techniques have also been developed fairly extensively. When compared with speech coding, a clear difference is that the systems must have access to large databases of natural language and/or speech in order to obtain good results. Speech synthesis and recognition thus require that the computer has knowledge of the recognized or synthesized language at some level, something that is not needed in speech coding. — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 2277

Categorization

• DEFINE: Cleanly draws the conceptual line between coding (no language knowledge) versus synthesis/recognition (language knowledge required).