Bouts of vocalizations given by seven red deer stags were recorded over the rutting period, and homomorphic analysis and hidden Markov models (two techniques typically used for the automatic recognition of human speech utterances) were used to investigate whether the spectral envelope of the calls was individually distinctive. Bouts of common roars (the most common call type) were highly individually distinctive, with an average recognition percentage of 93.5%. A “temporal” split-sample approach indicated that although in most individuals these identity cues held over the rutting period, the ability of the models trained with the bouts of roars recorded early in the rut to correctly classify later vocalizations decreased as the recording date increased. When Markov models trained using the bouts of common roars were used to classify other call types according to their individual membership, the classification results indicated that the cues to identity contained in the common roars were also present in the other call types. This is the first demonstration in mammals other than primates that individuals have vocal cues to identity that are common to the different call types that compose their vocal repertoire.
17 August 2022 in EURASIP Journal on Audio, Speech, and Music Processing
Most state-of-the-art speech systems use deep neural networks (DNNs). These systems require a large amount of data to be learned. Hence, training state-of-the-art frameworks on under-resourced speech challenges are difficult tasks. As an example, a challenge could be the limited amount of data to model impaired speech. Furthermore, acquiring more data and/or expertise is time-consuming and expensive. In this paper, we focus on the following speech processing tasks: automatic speech recognition, speaker identification, and emotion recognition. To assess the problem of limited data, we firstly investigate state-of-the-art automatic speech recognition systems, as this is the hardest task (due to the wide variability in each language). Next, we provide an overview of techniques and tasks requiring fewer data. In the last section, we investigate few-shot techniques by interpreting under-resourced speech as a few-shot problem. In that sense, we propose an overview of few-shot techniques and the possibility of using such techniques for the speech problems addressed in this survey. It is true that the reviewed techniques are not well adapted for large datasets. Nevertheless, some promising results from the literature encourage the usage of such techniques for speech processing.
Background: The development of automatic tools based on acoustic analysis allows to overcome the limitations of perceptual assessment for patients with head and neck cancer. The aim of this study is to provide a systematic review of literature describing the effects of oral and oropharyngeal cancer on speech intelligibility using acoustic analysis.Methods: Two databases (PubMed and Embase) were surveyed. The selection process, according to the preferred reporting items for systematic reviews and meta‐analyses (PRISMA) statement, led to a final set of 22 articles.Results: Nasalance is studied mainly in oropharyngeal patients. The vowels are mostly studied using formant analysis and vowel space area, the consonants by means of spectral moments with specific parameters according to their phonetic characteristic. Machine learning methods allow classifying “intelligible” or “unintelligible” speech for T3 or T4 tumors.Conclusions: The development of comprehensive models combining different acoustic measures would allow a better consideration of the functional impact of the speech disorder.
18 September 2017 in Journal of Speech, Language, and Hearing Research
Purpose The purpose of this article is to assess speech processing for listeners with simulated age-related hearing loss (ARHL) and to investigate whether the observed performance can be replicated using an automatic speech recognition (ASR) system. The long-term goal of this research is to develop a system that will assist audiologists/hearing-aid dispensers in the fine-tuning of hearing aids. Method Sixty young participants with normal hearing listened to speech materials mimicking the perceptual consequences of ARHL at different levels of severity. Two intelligibility tests (repetition of words and sentences) and 1 comprehension test (responding to oral commands by moving virtual objects) were administered. Several language models were developed and used by the ASR system in order to fit human performances. Results Strong significant positive correlations were observed between human and ASR scores, with coefficients up to .99. However, the spectral smearing used to simulate losses in frequency selectivity caused larger declines in ASR performance than in human performance. Conclusion Both intelligibility and comprehension scores for listeners with simulated ARHL are highly correlated with the performances of an ASR-based system. In the future, it needs to be determined if the ASR system is similarly successful in predicting speech processing in noise and by older people with ARHL.