Mercurial > hg > tony
comparison publications/sempre2014/mauch_sempre2014_GF_edits.txt @ 175:26224791546f
edited sempre paper
author | gyorgyf |
---|---|
date | Sun, 09 Feb 2014 12:58:28 +0000 |
parents | |
children |
comparison
equal
deleted
inserted
replaced
174:e33f9d052503 | 175:26224791546f |
---|---|
1 Paper title. | |
2 Matthias Mauch and Chris Cannam: Efficient Computer-Aided Pitch Track and Note Estimation for Scientific Applications | |
3 | |
4 Abstract. | |
5 We present **Tony**, a free, open-source software tool for | |
6 computer-aided pitch track and note annotation of melodic audio content. | |
7 The accurate annotation of fundamental frequencies and notes | |
8 is essential to the scientific study of | |
9 intonation in singing and other instruments. | |
10 Unlike commercial applications for singers and producers | |
11 or other academic tools for generic music annotation and visualisation | |
12 **Tony** has been designed for the scientific study of monophonic music: | |
13 a) it implements state-of-the art algorithms for pitch and note estimation from audio, | |
14 b) it provides visual and auditory feedback of the extracted pitches | |
15 for the identification of detection errors, | |
16 b) it provides an intelligent graphical user interface | |
17 through which the user can identify and rapidly correct estimation errors, | |
18 c) it provides functions for exporting pitch track and note track | |
19 enabling further processing in spreadsheets or other applications. | |
20 Software versions for Windows, OSX and Linux platforms can be downloaded from | |
21 http://code.soundsoftware.ac.uk/projects/tony | |
22 | |
23 Keyword 1. | |
24 Pitch/Note Analysis | |
25 | |
26 Keyword 2. | |
27 Software | |
28 | |
29 Keyword 3. | |
30 Singing. | |
31 | |
32 Aims. | |
33 We aim to make the scientific annotation of melodic content more efficient. | |
34 ==> We aim to make the annotation of melodic content for scientific purposes more efficient. | |
35 (also, possibly move this sentence to the end) | |
36 | |
37 Music psychologists interested in the analysis of pitch and intonation | |
38 usually use software programs originally aimed at the analysis of speech | |
39 (e.g. Praat http://www.fon.hum.uva.nl/praat/) or generic audio annotation | |
40 tools (e.g. Sonic Visualiser http://www.sonicvisualiser.org/) | |
41 to extract pitches of notes from audio recordings. | |
42 Since these programs were not conceived for musical pitch analysis, | |
43 the process of extracting note frequencies remains laborious and can take | |
44 many times the duration of the recording. | |
45 | |
46 On the other hand, commercial tools such as | |
47 Melodyne (http://www.celemony.com/), Songs2See (http://www.songs2see.com/) or | |
48 Sing&See (http://www.singandsee.com/) have | |
49 unknown frequency estimation procedures (proprietary code) | |
50 and do not provide export formats needed for scientific analysis. | |
51 | |
52 ==> Commercial tools such as Melodyne (http://www.celemony.com/), Songs2See (http://www.songs2see.com/) or | |
53 Sing&See (http://www.singandsee.com/) also exists for these purposes, however | |
54 their frequency estimation procedures are typically not public (proprietary code), | |
55 and they do not provide export formats suitable for scientific analysis. | |
56 | |
57 | |
58 An academic note annotation system [1] exists, but does not feature | |
59 note extraction. It is also not openly available. | |
60 | |
61 ==> An note annotation system [1] developed for academic purposes exists, but it does not feature | |
62 note extraction. It is also not openly available. (openly ?? => open source, free/prop.? ) | |
63 | |
64 This is why, during our own research on intonation [2], | |
65 we decided to code our own pitch extraction tool that would avoid the shortcomings. | |
66 | |
67 ==> This is why we decided to develop our own pitch extraction tool that would avoid | |
68 the above shortcomings during our own research on intonation [2]. | |
69 | |
70 | |
71 Methods. | |
72 For automatic pitch and note estimation we use the pYIN method [3]. | |
73 The method provides precise pitch and note estimates and | |
74 automatically determines which parts of the recording are voiced. | |
75 | |
76 The graphical user interface is based upon the | |
77 open source software libraries from Sonic Visualiser. | |
78 | |
79 ==> The graphical user interface is based upon | |
80 open source software libraries originally developed for the Sonic Visualiser software. | |
81 | |
82 It features the audio waveform, a spectrogram representation, | |
83 the pitch track and notes. Users can scroll and zoom in time. | |
84 **Tony** does not only play back the original audio, | |
85 but also, optionally, sonifications of the pitch track (melody line) | |
86 and the note track (discrete pitches with durations). | |
87 Notes' pitches are robustly estimated as the median of the pitch track | |
88 that occurs during the duration of the note. | |
89 | |
90 (robustly? I know it's good, but nothing really supports the fairly strong statement hereā¦) | |
91 | |
92 The user can delete, move, cut, merge, crop and extend notes, | |
93 and the note's frequency is adapted accordingly. | |
94 The user can delete spurious parts of the pitch track | |
95 and shift the pitch track in frequency. | |
96 In order to efficiently correct erroneous pitch tracks, the user can select | |
97 a time interval, and **Tony** will provide various alternative | |
98 pitch tracks. The user can then pick the correct one. | |
99 | |
100 Outcomes. | |
101 The system is currently being used for two projects: | |
102 for the generation of new training and test data for Music Informatics research, | |
103 and for a new project on intonation in unaccompanied solo singing. | |
104 | |
105 ==> The system is currently being used for two projects: | |
106 1) for the generation of new training and test data for Music Informatics research, | |
107 and 2) a research project on intonation in unaccompanied solo singing. | |
108 | |
109 Preliminary feedback by the users suggests that | |
110 the system does indeed facilitate pitch annotation | |
111 and provides vital features that cannot be found in other tools. | |
112 | |
113 | |
114 Title for final section. | |
115 Conclusions | |
116 | |
117 [Q37]. | |
118 We presented **Tony** a new software tool for computer-assisted | |
119 annotation of melodic audio content for scientific analysis. | |
120 No other existing program combines pitch and note estimation, | |
121 a graphical user interface with auditory feedback, | |
122 rapid, computer-aided correction of pitches and | |
123 and extensive exporting facilities. | |
124 **Tony** is freely available for use on Windows, OSX and Linux platforms | |
125 from http://code.soundsoftware.ac.uk/projects/tony/. | |
126 | |
127 Acknowledgements. | |
128 Matthias Mauch is funded by the Royal Academy of Engineering. | |
129 We would like to thank Justin Salamon, Rachel Bittner and Juan Bello | |
130 for their comments and coding help. | |
131 | |
132 Three key references. (APA v6) | |
133 [1] Pant, S., Rao, V., & Rao, P. (2010). A melody detection user interface for polyphonic music. 2010 National Conference On Communications (NCC), 2010. | |
134 [2] Mauch, M., Frieler, K., & Dixon, S. (under review). Intonation in Unaccompanied Singing: Accuracy, Drift and a Model of Intonation Memory. | |
135 [3] Mauch, M., & Dixon, S. (2014). pYIN : a Fundamental Frequency Estimator Using Probabilistic Threshold Distributions. In Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2014). | |
136 | |
137 Comments/queries to organisers. | |
138 |