Advances in computer science and digital speech processing increasingly allow the development of programs and tools that allow the performance of tasks related to the analysis and synthesis of speech automatically or semi-automatically.
This line of research aims to develop automatic speech processing tools that can be applied to both analysis and automatic speech generation. The aim is to have tools that allow the processing of large corpora of speech completely automatically, with a level of quality high enough to use its output without the need for subsequent manual review, or with minimal review.
In contrast to other approaches that explore the use of statistical techniques and machine learning for the development of such tools, the direct application of existing linguistic / phonetic knowledge to the development of heuristic systems is intended here, which make serve rules.
This line has so far resulted in the development of a number of tools for speech analysis and synthesis: