Comparing the Accuracy of Visual and Computerized Onset Detection Methods on Simulated Electromyography Signals with Varying Signal-to-Noise Ratios

Electromyography (EMG) onsets determined by computerized detection methods have been compared against the onsets selected by experts through visual inspection. However, with this type of approach, the true onset remains unknown, making it impossible to determine if computerized detection methods are better than visual detection (VD) as they can only be as good as what the experts select. The use of simulated signals allows for all aspects of the signal to be precisely controlled, including the onset and the signal-to-noise ratio (SNR). This study compared three onset detection methods: approximated generalized likelihood ratio, double threshold (DT), and VD determined by eight trained individuals. The selected onset was compared against the true onset in simulated signals which varied in the SNR from 5 to 40 dB. For signals with 5 dB SNR, the VD method was significantly better, but for SNRs of 20 dB or greater, no differences existed between the VD and DT methods. The DT method is recommended as it can improve objectivity and reduce time of analysis when determining EMG onsets. Even for the best-quality signals (SNR of 40 dB), all the detection methods were off by 15–30 ms from the true onset and became progressively more inaccurate as the SNR decreased. Therefore, although all the detection methods provided similar results, they can be off by 50–80 ms from the true onset as the SNR decreases to 10 dB. Caution must be used when interpreting EMG onsets, especially on signals where the SNR is low or not reported at all.

COSMO-Onset: A Neurally-Inspired Computational Model of Spoken Word Recognition, Combining Top-Down Prediction and Bottom-Up Detection of Syllabic Onsets

Recent neurocognitive models commonly consider speech perception as a hierarchy of processes, each corresponding to specific temporal scales of collective oscillatory processes in the cortex: 30–80 Hz gamma oscillations in charge of phonetic analysis, 4–9 Hz theta oscillations in charge of syllabic segmentation, 1–2 Hz delta oscillations processing prosodic/syntactic units and the 15–20 Hz beta channel possibly involved in top-down predictions. Several recent neuro-computational models thus feature theta oscillations, driven by the speech acoustic envelope, to achieve syllabic parsing before lexical access. However, it is unlikely that such syllabic parsing, performed in a purely bottom-up manner from envelope variations, would be totally efficient in all situations, especially in adverse sensory conditions. We present a new probabilistic model of spoken word recognition, called COSMO-Onset, in which syllabic parsing relies on fusion between top-down, lexical prediction of onset events and bottom-up onset detection from the acoustic envelope. We report preliminary simulations, analyzing how the model performs syllabic parsing and phone, syllable and word recognition. We show that, while purely bottom-up onset detection is sufficient for word recognition in nominal conditions, top-down prediction of syllabic onset events allows overcoming challenging adverse conditions, such as when the acoustic envelope is degraded, leading either to spurious or missing onset events in the sensory signal. This provides a proposal for a possible computational functional role of top-down, predictive processes during speech recognition, consistent with recent models of neuronal oscillatory processes.

Musical note onset detection based on a spectral sparsity measure

If music is the language of the universe, musical note onsets may be the syllables for this language. Not only do note onsets define the temporal pattern of a musical piece, but their time-frequency characteristics also contain rich information about the identity of the musical instrument producing the notes. Note onset detection (NOD) is the basic component for many music information retrieval tasks and has attracted significant interest in audio signal processing research. In this paper, we propose an NOD method based on a novel feature coined as Normalized Identification of Note Onset based on Spectral Sparsity (NINOS2). The NINOS2 feature can be thought of as a spectral sparsity measure, aiming to exploit the difference in spectral sparsity between the different parts of a musical note. This spectral structure is revealed when focusing on low-magnitude spectral components that are traditionally filtered out when computing note onset features. We present an extensive set of NOD simulation results covering a wide range of instruments, playing styles, and mixing options. The proposed algorithm consistently outperforms the baseline Logarithmic Spectral Flux (LSF) feature for the most difficult group of instruments which are the sustained-strings instruments. It also shows better performance for challenging scenarios including polyphonic music and vibrato performances.