Underwater Signal Analysis in the Modulation Spectrogram with Time-Frequency Reassignment Technique

Cho, Hyunjin; Kim, Wan Jin; Hong, Wooyoung

doi:10.1587/transfun.e102.a.1542

Cited by 4 publications

(1 citation statement)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reassignment techniques have also been used to improve the analysis of wolf howls [36] and to develop a TFR tailored for the study of underwater signals [23]. 15 Mel-Frequency cepstral coefficients.…”

Section: Reassignment and Multi-taperingmentioning

confidence: 99%

Bioacoustics through the Time-Frequency plane

Listanti

View full text Add to dashboard Cite

The natural world is full of sounds produced by plants, animals and even landscape elements. Animals can be incredibly creative in their sound production, and the study of their vocalisation is an essential part of behavioural and conservation ecology. Counting animals' vocalisation is often used as a survey method to monitor population abundances, especially with cryptic or nocturnal species. Once performed only by humans, in the last decades, call counts have evolved, and nowadays, Automatic Recording Units (ARUs) are often used for this purpose, and terabytes of data are collected during acoustics surveys requiring long hours of tedious work to be analysed.As a consequence, in recent years, there has been a growing interest in using signal processing and artificial intelligence methods to speed up and facilitate this process.Studying sounds, even animal sounds, entails studying how their frequency and intensity change in time, a piece of information not easily readable from the data type collected by ARUs: the waveform. A waveform only describes the change of amplitude with respect to time, but other analysis tools are needed to have information about the instantaneous changes in frequency and intensity.In 1946, Koening, Dunn and Lace introduced a 3D representation of sounds, where time, frequency and intensity could be read simultaneously: the spectrogram. The spectrogram, then, became one of the main tools used in the broad field of signal processing and, more specifically, in the study of animal and natural sounds: Bioacoustics. Because the spectrogram combines the representation of a signal in both the time and frequency domain, it is called a Time-Frequency Representation (TFR) of sound.However, during these seven decades, more TFRs were introduced to improve the spectrogram, whose performances are hindered by its limits in the time-frequency resolution. Every TFR is defined by a set of parameters that can influence its ability to depict important sound features effectively. Therefore, choosing the best TFR and the best parameters is a challenging task, and it is highly dependent on the characteristics of the sound we are studying. In this thesis, we will explore the differences between the main TFRs present in literature, and we will propose methods to test their performance in some real-world bioacoustics problems involving Aotearoa/New Zealand bats' echolocation data and North Island Brown kiwi calls data. We will also demonstrate how the choice of a TFR and its parameters is crucial if we want to obtain optimal results from our data analysis and discuss the importance of TFRs in Bioacoustics.

show abstract

Section: Reassignment and Multi-taperingmentioning

confidence: 99%

Bioacoustics through the Time-Frequency plane

Listanti

View full text Add to dashboard Cite

show abstract

Formants are easy to measure; resonances, not so much: Lessons from Klatt (1986)

Whalen

Chen

Shadle

et al. 2022

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

Formants in speech signals are easily identified, largely because formants are defined to be local maxima in the wideband sound spectrum. Sadly, this is not what is of most interest in analyzing speech; instead, resonances of the vocal tract are of interest, and they are much harder to measure. Klatt [(1986). in Proceedings of the Montreal Satellite Symposium on Speech Recognition, 12th International Congress on Acoustics, edited by P. Mermelstein (Canadian Acoustical Society, Montreal), pp. 5–7] showed that estimates of resonances are biased by harmonics while the human ear is not. Several analysis techniques placed the formant closer to a strong harmonic than to the center of the resonance. This “harmonic attraction” can persist with newer algorithms and in hand measurements, and systematic errors can persist even in large corpora. Research has shown that the reassigned spectrogram is less subject to these errors than linear predictive coding and similar measures, but it has not been satisfactorily automated, making its wider use unrealistic. Pending better techniques, the recommendations are (1) acknowledge limitations of current analyses regarding influence of F0 and limits on granularity, (2) report settings more fully, (3) justify settings chosen, and (4) examine the pattern of F0 vs F1 for possible harmonic bias.

show abstract

Assessing accuracy of resonances obtained with reassigned spectrograms from the “ground truth” of physical vocal tract models

Shadle,

Fulop,

Chen

et al. 2024

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

The reassigned spectrogram (RS) has emerged as the most accurate way to infer vocal tract resonances from the acoustic signal [Shadle, Nam, and Whalen (2016). “Comparing measurement errors for formants in synthetic and natural vowels,” J. Acoust. Soc. Am. 139(2), 713–727]. To date, validating its accuracy has depended on formant synthesis for ground truth values of these resonances. Synthesis is easily controlled, but it has many intrinsic assumptions that do not necessarily accurately realize the acoustics in the way that physical resonances would. Here, we show that physical models of the vocal tract with derivable resonance values allow a separate approach to the ground truth, with a different range of limitations. Our three-dimensional printed vocal tract models were excited by white noise, allowing an accurate determination of the resonance frequencies. Then, sources with a range of fundamental frequencies were implemented, allowing a direct assessment of whether RS avoided the systematic bias towards the nearest strong harmonic to which other analysis techniques are prone. RS was indeed accurate at fundamental frequencies up to 300 Hz; above that, accuracy was somewhat reduced. Future directions include testing mechanical models with the dimensions of children's vocal tracts and making RS more broadly useful by automating the detection of resonances.

show abstract

Underwater Signal Analysis in the Modulation Spectrogram with Time-Frequency Reassignment Technique

Cited by 4 publications

References 3 publications

Bioacoustics through the Time-Frequency plane

Bioacoustics through the Time-Frequency plane

Formants are easy to measure; resonances, not so much: Lessons from Klatt (1986)

Assessing accuracy of resonances obtained with reassigned spectrograms from the “ground truth” of physical vocal tract models

Contact Info

Product

Resources

About