Refine
Document Type
- Article (1)
- Conference Proceeding (1)
Language
- English (2)
Is part of the Bibliography
- yes (2)
Keywords
- Audio Desciptors (1)
- Klassische Musik (1)
- Lispeln (1)
- Musikaufnahme (1)
- Raumakustik (1)
- Räumliches Hören (1)
- Signalverarbeitung (1)
- räumliche Wahrnehmung (1)
Department/institution
Correcting lisps in speech can prove to be of great difficulty to many, as they may be unaware of whether they are lisping. To help those affected, we have developed a simple algorithm for the real-time identification of the sigmatismus lateralis in ”S” sounds within speech via analysis in the frequency domain. The algorithm identifies peaks within the lisp’s frequency band after calibration. A frequency band of 3000-4000 Hz has been identified to be generally accurate for lisp and 2500-3000 Hz for the correct pronunciation for a single male test subject. The algorithm splits given speech recordings into smaller segments and compares the number of lisps and non-lisps detected in these segments to categorize. From tests, it was concluded that a segment length of 0.5 s produces the best results. The algorithm does not detect every lisp section, however it does not raise false positives. Our implementation in Julia with multi-threaded per-file analysis is able to analyze 20 files of lengths between 5 s and 10 s within 0.21 s on a Qualcomm Snapdragon 860 smartphone chipset, meaning analysis is far faster than real-time. The proposed algorithm is a simple prototype algorithm capable of realtime analysis of audio in the frequency domain to identify whether lateral lisps are the dominant sibilant pronunciation in a given window. The method was only tested for a single test subject. However, a calibration algorithm capable of adjusting parameters to new individuals is proposed. The algorithm itself should be easily expandable to identify other speech impediments.
The perceived spatial impression is an important factor in the aesthetic appreciation of classical music recordings. Crucial aspects of spatial perception can be described by the attributes ensemble width, depth, envelopment, and reverberation duration. In a listening experiment, a sample of 30 experts rated 24 orchestra recordings by these attributes. To compare these ratings to signalrelated measures of spatial impression, an audio descriptor for the interaural cross-correlation coefficient in music recordings was chosen as a measure of perceived ensemble width. A second algorithm estimated the reverberation time from decay segments in the recorded music. Linear mixed-effects models were then used to test whether these two descriptors, combined with other audio features related to timbre, are suitable for predicting participants’ ratings of the perceptual attributes. Overall, results confirmed that the signal-related measures used can indeed predict perceptual attributes, with 41% of rating variance explained in the case of reverberation duration. In contrast, these objective measures turned out to be less accurate predictors of the perceived ensemble width, with a variance explanation of 21%. This study can thus be regarded as a starting point for further studies aiming to predict the perceived spatial impression in music recordings.