| 1. |
- Benetos, Emmanouil, et al.
(författare)
-
Automatic Transcription of Turkish Makam Music
- 2013
-
Ingår i: Proceedings of ISMIR - International Conference on Music Information Retrieval. - : International Society for Music Information Retrieval. ; , s. 355-360
-
Konferensbidrag (refereegranskat)abstract
- In this paper we propose an automatic system for transcribing makam music of Turkey. We document the specific traits of this music that deviate from properties that were targeted by transcription tools so far and we compile a dataset of makam recordings along with aligned microtonal ground-truth. An existing multi-pitch detection algorithm is adapted for transcribing music in 20 cent resolution, and the final transcription is centered around the tonic frequency of the recording. Evaluation metrics for transcribing microtonal music are utilized and results show that transcription of Turkish makam music in e.g. an interactive transcription software is feasible using the current state-of-the-art.
|
|
| 2. |
- Benetos, Emmanouil, et al.
(författare)
-
Automatic transcription of Turkish microtonal music
- 2015
-
Ingår i: Journal of the Acoustical Society of America. - : Acoustical Society of America (ASA). - 0001-4966. ; 138:4, s. 2118-2130
-
Tidskriftsartikel (refereegranskat)abstract
- Automatic music transcription, a central topic in music signal analysis, is typically limited to equal-tempered music and evaluated on a quartertone tolerance level. A system is proposed to automatically transcribe microtonal and heterophonic music as applied to the makam music of Turkey. Specific traits of this music that deviate from properties targeted by current transcription tools are discussed, and a collection of instrumental and vocal recordings is compiled, along with aligned microtonal reference pitch annotations. An existing multi-pitch detection algorithm is adapted for transcribing music with 20 cent resolution, and a method for converting a multi-pitch heterophonic output into a single melodic line is proposed. Evaluation metrics for transcribing microtonal music are applied, which use various levels of tolerance for inaccuracies with respect to frequency and time. Results show that the system is able to transcribe microtonal instrumental music at 20 cent resolution with an F-measure of 56.7%, outperforming state-of-the-art methods for the same task. Case studies on transcribed recordings are provided, to demonstrate the shortcomings and the strengths of the proposed method.
|
|
| 3. |
- Benetos, Emmanouil, et al.
(författare)
-
Incorporating pitch class profiles for improving automatic transcription of Turkish makam music
- 2014
-
Ingår i: Proceedings of the 4th Workshop on Folk Music Analysis. - : Computer Engineering Department, Bogaziçi University. ; , s. 15-20
-
Konferensbidrag (refereegranskat)abstract
- In this paper we evaluate the impact of including knowledge about scale material into a system for the transcription of Turkish makam music. To this end, we extend our previously presented appoach by a refinement iteration that gives preference to note values present in the scale of the mode (i.e. makam). The information about the scalar material is provided in form of pitch class profiles, and they are imposed in form of a Dirichlet prior to our expanded probabilistic latent component analysis (PLCA) transcription system. While the inclusion of such a prior was supposed to focus the transcription system on musically meaningful areas, the obtained results are significantly improved only for recordings of certain instruments. In our discussion we demonstrate the quality of the obtained transcriptions, and discuss the difficulties caused for evaluation in the context of microtonal music.
|
|
| 4. |
- Benetos, Emmanouil, et al.
(författare)
-
Pitched Instrument Onset Detection Based on Auditory Spectra
- 2009
-
Ingår i: Proceedings of ISMIR - International Conference on Music Information Retrieval. ; , s. 105-110
-
Konferensbidrag (refereegranskat)abstract
- In this paper, a novel method for onset detection of music signals using auditory spectra is proposed. The auditory spectrogram provides a time-frequency representation that employs a sound processing model resembling the human auditory system. Recent work on onset detection employs DFT-based features, such as the spectral flux and group delay function. The spectral flux and group delay are introduced in the auditory framework and an onset detection algorithm is proposed. Experiments are conducted on a dataset covering 11pitched instrument types, consisting of 1829 onsets in total. Results indicate the superiority of the auditory representations over the DFT-based ones, with the auditory spectral flux exhibiting an onset detection improvement by 2% in terms of F-measure when compared to the DFT-based feature.
|
|
| 5. |
- Chettri, Bhusan, et al.
(författare)
-
ANALYSING REPLAY SPOOFING COUNTERMEASURE PERFORMANCE UNDER VARIED CONDITIONS
- 2018
-
Ingår i: 2018 IEEE 28TH INTERNATIONAL WORKSHOP ON MACHINE LEARNING FOR SIGNAL PROCESSING (MLSP). - : IEEE. - 9781538654774
-
Konferensbidrag (refereegranskat)abstract
- In this paper, we aim to understand what makes replay spoofing detection difficult in the context of the ASVspoof 2017 corpus. We use FFT spectra, mel frequency cepstral coefficients (MFCC) and inverted MFCC (IMFCC) frontends and investigate different back-ends based on Convolutional Neural Networks (CNNs), Gaussian Mixture Models (GMMs) and Support Vector Machines (SVMs). On this database, we find that IMFCC frontend based systems show smaller equal error rate (EER) for high quality replay attacks but higher EER for low quality replay attacks in comparison to the baseline. However, we find that it is not straightforward to understand the influence of an acoustic environment (AE), a playback device (PD) and a recording device (RD) of a replay spoofing attack. One reason is the unavailability of metadata for genuine recordings. Second, it is difficult to account for the effects of the factors: AE, PD and RD, and their interactions. Finally, our frame-level analysis shows that the presence of cues (recording artefacts) in the first few frames of genuine signals (missing from replayed ones) influence class prediction.
|
|
| 6. |
- Chettri, Bhusan, et al.
(författare)
-
Dataset Artefacts in Anti-Spoofing Systems : A Case Study on the ASVspoof 2017 Benchmark
- 2020
-
Ingår i: IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING. - : Institute of Electrical and Electronics Engineers (IEEE). - 2329-9290. ; 28, s. 3018-3028
-
Tidskriftsartikel (refereegranskat)abstract
- The Automatic Speaker Verification Spoofing and Countermeasures Challenges motivate research in protecting speech biometric systems against a variety of different access attacks. The 2017 edition focused on replay spoofing attacks, and involved participants building and training systems on a provided dataset (ASVspoof 2017). More than 60 research papers have so far been published with this dataset, but none have sought to answer why countermeasures appear successful in detecting spoofing attacks. This article shows how artefacts inherent to the dataset may be contributing to the apparent success of published systems. We first inspect the ASVspoof 2017 dataset and summarize various artefacts present in the dataset. Second, we demonstrate how countermeasure models can exploit these artefacts to appear successful in this dataset. Third, for reliable and robust performance estimates on this dataset we propose discarding nonspeech segments and silence before and after the speech utterance during training and inference. We create speech start and endpoint annotations in the dataset and demonstrate how using them helps countermeasure models become less vulnerable from being manipulated using artefacts found in the dataset. Finally, we provide several new benchmark results for both frame-level and utterance-level models that can serve as new baselines on this dataset.
|
|
| 7. |
- Holzapfel, Andre, et al.
(författare)
-
Automatic music transcription and ethnomusicology : A user study
- 2019
-
Ingår i: Proceedings of the 20th International Society for Music Information Retrieval Conference, ISMIR 2019. ; , s. 678-684
-
Konferensbidrag (refereegranskat)abstract
- Converting an acoustic music signal into music notation using a computer program has been at the forefront of music information research for several decades, as a task referred to as automatic music transcription (AMT). However, current AMT research is still constrained to system development followed by quantitative evaluations; it is still unclear whether the performance of AMT methods is considered sufficient to be used in the everyday practice of music scholars. In this paper, we propose and carry out a user study on evaluating the usefulness of automatic music transcription in the context of ethnomusicology. As part of the study, we recruited 16 participants who were asked to transcribe short musical excerpts either from scratch or using the output of an AMT system as a basis. We collect and analyze quantitative measures such as transcription time and effort, and a range of qualitative feedback from study participants, which includes user needs, criticisms of AMT technologies, and links between perceptual and quantitative evaluations on AMT outputs. The results show no quantitative advantage of using AMT, but important indications regarding appropriate user groups and evaluation measures are provided.
|
|
| 8. |
- Holzapfel, Andre, et al.
(författare)
-
Humanities and engineering perspectives on music transcription
- 2021
-
Ingår i: Digital Scholarship in the Humanities. - : Oxford University Press (OUP). - 2055-7671 .- 2055-768X.
-
Tidskriftsartikel (refereegranskat)abstract
- Music transcription is a process of creating a notation of musical sounds. It has been used as a basis for the analysis of music from a wide variety of cultures. Recent decades have seen an increasing amount of engineering research within the field of Music Information Retrieval that aims at automatically obtaining music transcriptions in Western staff notation. However, such approaches are not widely applied in research in ethnomusicology. This article aims to bridge interdisciplinary gaps by identifying aspects of proximity and divergence between the two fields. As part of our study, we collected manual transcriptions of traditional dance tune recordings by eighteen transcribers. Our method employs a combination of expert and computational evaluation of these transcriptions. This enables us to investigate the limitations of automatic music transcription (AMT) methods and computational transcription metrics that have been proposed for their evaluation. Based on these findings, we discuss promising avenues to make AMT more useful for studies in the Humanities. These are, first, assessing the quality of a transcription based on an analytic purpose; secondly, developing AMT approaches that are able to learn conventions concerning the transcription of a specific style; thirdly, a focus on novice transcribers as users of AMT systems; and, finally, considering target notation systems different from Western staff notation.
|
|
| 9. |
- Holzapfel, André, 1976-, et al.
(författare)
-
The Sousta corpus : Beat-informed automatic transcription of traditional dance tunes
- 2016
-
Ingår i: Proceedings of ISMIR - International Conference on Music Information Retrieval. ; , s. 531-537
-
Konferensbidrag (refereegranskat)abstract
- In this paper, we present a new corpus for research in computational ethnomusicology and automatic music transcription, consisting of traditional dance tunes from Crete. This rich dataset includes audio recordings, scores transcribed by ethnomusicologists and aligned to the audio performances, and meter annotations. A second contribution of this paper is the creation of an automatic music transcription system able to support the detection of multiple pitches produced by lyra (a bowed string instrument). Furthermore, the transcription system is able to cope with deviations from standard tuning, and provides temporally quantized notes by combining the output of the multi-pitch detection stage with a state-of-the-art meter tracking algorithm. Experiments carried out for note tracking using 25ms onset tolerance reach 41.1% using information from the multi-pitch detection stage only, 54.6% when integrating beat information, and 57.9% when also supporting tuning estimation. The produced meter aligned transcriptions can be used to generate staff notation, a fact that increases the value of the system for studies in ethnomusicologyIn this paper, we present a new corpus for research in computational ethnomusicology and automatic music transcription, consisting of traditional dance tunes from Crete. This rich dataset includes audio recordings, scores transcribed by ethnomusicologists and aligned to the audio performances, and meter annotations. A second contribution of this paper is the creation of an automatic music transcription system able to support the detection of multiple pitches produced by lyra (a bowed string instrument). Furthermore, the transcription system is able to cope with deviations from standard tuning, and provides temporally quantized notes by combining the output of the multi-pitch detection stage with a state-of-the-art meter tracking algorithm. Experiments carried out for note tracking using 25ms onset tolerance reach 41.1% using information from the multi-pitch detection stage only, 54.6% when integrating beat information, and 57.9% when also supporting tuning estimation. The produced meter aligned transcriptions can be used to generate staff notation, a fact that increases the value of the system for studies in ethnomusicology
|
|
| 10. |
|
|
