digital-garden-anabasis/pages/incubation.audio.synthesis.concatenative.md

# Concatenative synthesis

## 101
> **Concatenative synthesis** is a technique for synthesising sounds by concatenating short samples of recorded sound (called _units_).
>  - https://en.wikipedia.org/wiki/Concatenative_synthesis

## Adjacent
- [[incubation.audio.sample.managment]]
- [[tools.tts]]


## Tools
### Unsorted
- https://en.wikipedia.org/wiki/Festival_Speech_Synthesis_System
- https://en.wikipedia.org/wiki/ESpeak

	- - https://www.isi.edu/~carte/e-speech/synth/index.html


- https://en.wikipedia.org/wiki/Sinsy
	- > **Sinsy** (**Sin**ging Voice **Sy**nthesis System) (しぃんしぃ) is an online [Hidden Markov model](https://en.wikipedia.org/wiki/Hidden_Markov_model "Hidden Markov model") (HMM)-based singing voice synthesis system by the [Nagoya Institute of Technology](https://en.wikipedia.org/wiki/Nagoya_Institute_of_Technology "Nagoya Institute of Technology") that was created under the [Modified BSD license](https://en.wikipedia.org/wiki/BSD_licenses "BSD licenses").[](https://en.wikipedia.org/wiki/Sinsy#cite_note-1)
	- http://www.sinsy.jp/


- https://www.audiolabs-erlangen.de/resources/MIR/2015-ISMIR-LetItBee
	- http://labrosa.ee.columbia.edu/hamr_ismir2014/proceedings/doku.php?id=audio_mosaicing

- https://www.danieleghisi.com/phd/PHDThesis_20180118.pdf
- https://musicinformationretrieval.com/



### Promising
- https://colab.research.google.com/github/stevetjoa/musicinformationretrieval.com/blob/gh-pages/nmf_audio_mosaic.ipynb
- https://soundlab.cs.princeton.edu/research/mosievius/
- http://imtr.ircam.fr/imtr/Corpus_Based_Synthesis
	- max/msp
	- http://imtr.ircam.fr/imtr/Diemo_Schwarz
- https://github.com/benhackbarth/audioguide
	- OSX but promising
	
#### Audiostellar [8/10]
- https://audiostellar.xyz/
	- https://www.arj.no/tag/sox/
		- cataRT
	- ableton link incoming 



#### timbreIDLib Pure Data
https://github.com/wbrent/timbreIDLib
>timbreIDLib is a library of audio analysis externals for Pure Data. The classification external [timbreID] accepts arbitrary lists of audio features and attempts to find the best match between an input feature and previously stored training instances. The library can be used for a variety of real-time and non-real-time applications, including sound classification, sound searching, sound visualization, automatic segmenting, ordering of sounds by timbre, key and tempo estimation, and concatenative synthesis.

- https://forum.pdpatchrepo.info/topic/11876/scrambled-hackz-how-did-he-do-it