Speech Synthesis: Difference between revisions
Jump to navigation
Jump to search
No edit summary |
No edit summary |
||
| Line 3: | Line 3: | ||
* [[Neural network-based models]](1795-20s) | * [[Neural network-based models]](1795-20s) | ||
* [[Hidden Markov Models in Speech Synthesis]] (1980s-Now) Application of Hidden Markov Models in Multilingual Speech Synthesis： A Breakthrough in Text-to-Speech Systems. | * [[Hidden Markov Models in Speech Synthesis]] (1980s-Now) Application of Hidden Markov Models in Multilingual Speech Synthesis： A Breakthrough in Text-to-Speech Systems. | ||
===== New ideas ===== | |||
* [[Wolfgang von Kempelen's Speaking Machine (1769)]] | * [[Wolfgang von Kempelen's Speaking Machine (1769)]] | ||
* [[Wheatstone's "Speaking Machine" (1824)]] | * [[Wheatstone's "Speaking Machine" (1824)]] | ||
* [[Vocoder Development]] (1940s) | * [[Vocoder Development]] (1940s) | ||
Revision as of 11:57, 8 October 2023
- Early attempts at mechanical synthesis: Christian Kratzenstein (1779), Wolfgang van Kempelen ('Acoustic-Mechanical Speech Machine', 1791), Charles Wheatstone (1840s).
- The Bell Telephone Laboratory's Voder (1930s).
- Neural network-based models(1795-20s)
- Hidden Markov Models in Speech Synthesis (1980s-Now) Application of Hidden Markov Models in Multilingual Speech Synthesis： A Breakthrough in Text-to-Speech Systems.
New ideas
- Wheatstone's "Speaking Machine" (1824)
- Vocoder Development (1940s)
- The Parametric Artificial Talker (1952)
- IBM's "Audrey" (1956)
- Bell Labs' "Daisy Bell" (1968)
- Linear Predictive Coding (1970s)
- DECtalk Speech Synthesizer (1984)
- Carnegie Mellon's FestVox Project (1984)
- FestVox (1987)
- Festival Speech Synthesis System (1997)
- Advancements in Neural Network-Based TTS (2000s)
- Commerical TTS - Google, Amazon and Apple (2010s)
- Advancements in AI TTS (2020s)