Dataset of directional room impulse responses for realistic speech data
Ontology highlight
ABSTRACT:
SUBMITTER:
PROVIDER: S-EPMC10912595 | biostudies-literature | 2024 Feb
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Speech Emotion Recognition (SER) identifies and categorizes emotional states by analyzing speech signals. SER is an emerging research area using machine learning and deep learning techniques due to its socio-cultural and business importance. An appropriate dataset is an important resource for SER related studies in a particular language. There is an apparent lack of SER datasets in Bangla language although it is one of the most spoken languages in the world. There are a few Bangla SER datasets but those consist of only a few dialogs with a minimal number of actors making them unsuitable for real-world applications. Moreover, the existing datasets do not consider the intensity level of emotions. The intensity of a specific emotional expression, such as anger or sadness, plays a crucial role in social behavior. Therefore, a realistic Bangla speech dataset is developed in this study which is called KUET Bangla Emotional Speech (KBES) dataset. The dataset consists of 900 audio signals (i.e., speech dialogs) from 35 actors (20 females and 15 males) with diverse age ranges. Source of the speech dialogs are Bangla Telefilm, Drama, TV Series, Web Series. There are five emotional categories: Neutral, Happy, Sad, Angry, and Disgust. Except Neutral, samples of a particular emotion are divided into two intensity levels: Low and High. The significant issue of the dataset is that the speech dialogs are almost unique with relatively large number of actors; whereas, existing datasets (such as SUBESCO and BanglaSER) contain samples with repeatedly spoken of a few pre-defined dialogs by a few actors/research volunteers in the laboratory environment. Finally, the KBES dataset is exposed as a nine-class problem to classify emotions into nine categories: Neutral, Happy (Low), Happy (High), Sad (Low), Sad (High), Angry (Low), Angry (High), Disgust (Low) and Disgust (High). However, the dataset is kept symmetrical containing 100 samples for each of the nine classes; 100 samples are also gender balanced with 50 samples for male/female actors. The developed dataset seems a realistic dataset while compared with the existing SER datasets.
| S-EPMC10641593 | biostudies-literature
Project description:Room impulse responses (RIRs) are used in several applications, such as augmented reality and virtual reality. These applications require a large number of RIRs to be convolved with audio, under strict latency constraints. In this paper, we consider the compression of RIRs, in conjunction with fast time-domain convolution. We consider three different methods of RIR approximation for the purpose of RIR compression and compare them to state-of-the-art compression. The methods are evaluated using several standard objective quality measures, both channel-based and signal-based. We also propose a novel low-rank-based algorithm for fast time-domain convolution and show how the convolution can be carried out without the need to decompress the RIR. Numerical simulations are performed using RIRs of different lengths, recorded in three different rooms. It is shown that compression using low-rank approximation is a very compelling option to the state-of-the-art Opus compression, as it performs as well or better than on all but one considered measure, with the added benefit of being amenable to fast time-domain convolution.Supplementary informationThe online version contains supplementary material available at 10.1186/s13636-024-00363-5.
| S-EPMC11399173 | biostudies-literature
Project description:The Semantic Coherence Dataset has been designed to experiment with semantic coherence metrics. More specifically, the dataset has been built to the ends of testing whether probabilistic measures, such as perplexity, provide stable scores to analyze spoken language. Perplexity, which was originally conceived as an information-theoretic measure to assess the probabilistic inference properties of language models, has recently been proven to be an appropriate tool to categorize speech transcripts based on semantic coherence accounts. More specifically, perplexity has been successfully employed to discriminate subjects suffering from Alzheimer Disease and healthy controls. Collected data include speech transcripts, intended to investigate semantic coherence at different levels: data are thus arranged into two classes, to investigate intra-subject semantic coherence, and inter-subject semantic coherence. In the former case transcripts from a single speaker can be employed to train and test language models and to explore whether the perplexity metric provides stable scores in assessing talks from that speaker, while allowing to distinguish between two different forms of speech, political rallies and interviews. In the latter case, models can be trained by employing transcripts from a given speaker, and then used to measure how stable the perplexity metric is when computed using the model from that user and transcripts from different users. Transcripts were extracted from talks lasting almost 13 hours (overall 12:45:17 and 120,326 tokens) for the former class; and almost 30 hours (29:47:34 and 252,270 tokens) for the latter one. Data herein can be reused to perform analyses on measures built on top of language models, and more in general on measures that are aimed at exploring the linguistic features of text documents.
| S-EPMC9761592 | biostudies-literature
Project description: Not available
| S-EPMC11546432 | biostudies-literature
Project description:Deep Convolutional Neuronal Networks (DCNNs) are showing remarkable performance on many computer vision tasks. Due to their large parameter space, they require many labeled samples when trained in a supervised setting. The costs of annotating data manually can render the use of DCNNs infeasible. We present a novel framework called RenderGAN that can generate large amounts of realistic, labeled images by combining a 3D model and the Generative Adversarial Network framework. In our approach, image augmentations (e.g., lighting, background, and detail) are learned from unlabeled data such that the generated images are strikingly realistic while preserving the labels known from the 3D model. We apply the RenderGAN framework to generate images of barcode-like markers that are attached to honeybees. Training a DCNN on data generated by the RenderGAN yields considerably better performance than training it on various baselines.
| S-EPMC7805882 | biostudies-literature
Project description:This paper presents the Clarity Speech Corpus, a publicly available, forty speaker British English speech dataset. The corpus was created for the purpose of running listening tests to gauge speech intelligibility and quality in the Clarity Project, which has the goal of advancing speech signal processing by hearing aids through a series of challenges. The dataset is suitable for machine learning and other uses in speech and hearing technology, acoustics and psychoacoustics. The data comprises recordings of approximately 10,000 sentences drawn from the British National Corpus (BNC) with suitable length, words and grammatical construction for speech intelligibility testing. The collection process involved the selection of a subset of BNC sentences, the recording of these produced by 40 British English speakers, and the processing of these recordings to create individual sentence recordings with associated transcripts and metadata.
| S-EPMC8881678 | biostudies-literature
Project description:The most complex interactions between human beings occur through speech, and often in the presence of background noise. Understanding speech in noisy environments requires the integrity of highly integrated and widespread auditory networks likely to be impacted by multiple sclerosis (MS) related neurogenic injury. Despite the impact auditory communication has on a person's ability to navigate the world, build relationships, and maintain employability; studies of speech-in-noise (SiN) perception in people with MS (pwMS) have been minimal to date. Thus, this paper presents a dataset related to the acquisition of pure-tone thresholds, SiN performance and questionnaire responses in age-matched controls and pwMS. Bilateral pure-tone hearing thresholds were obtained at frequencies of 250 hertz (Hz), 500 Hz, 750 Hz, 1000 Hz, 1500 Hz, 2000 Hz, 4000 Hz, 6000 Hz and 8000 Hz, and hearing thresholds were defined as the lowest level at which the tone was perceived 50% of the time. Thresholds at 500 Hz, 1000 Hz, 2000 Hz and 4000 Hz were used to calculate the four-tone average for each participant, and only those with a bilateral four tone average of ≤ 25 dB HL were included in the analysis. To investigate SiN performance in pwMS, pre-recorded Bamford-Kowal-Bench (BKB) sentences were presented binaurally through headphones at five signal-to-noise ratios (SNR) in two noise conditions: speech-weighted noise and multi-talker babble. Participants were required to verbally repeat each sentence they had just heard; or indicate their inability to do so. A 33-item questionnaire, based on validated inventories for specific adult clinical populations with abnormal auditory processing, was used to evaluate auditory processing in daily life for pwMS. For analysis, pwMS were grouped according to their Expanded Disability Status Scale (EDSS) score as rated by a neurologist. PwMS with EDSS scores ≤ 1.5 were classified as 'mild' (n = 20); between 2 and 4.5 as 'moderate' (n = 16) and between 5 and 7 as 'advanced' (n = 10) and were compared to neurologically healthy controls (n = 38). The outcomes of the SiN task conducted in pwMS can be found in Iva et al., (2021). The present data has important implications for the timing and delivery of preparatory education to patients, family, and caregivers about communication abilities in pwMS. This dataset will also be valuable for the reuse/reanalysis required for future investigations into the clinical utility of SiN tasks to monitor disease progression.
| S-EPMC7726651 | biostudies-literature
Project description:BackgroundMicroarray technologies have become common tools in biological research. As a result, a need for effective computational methods for data analysis has emerged. Numerous different algorithms have been proposed for analyzing the data. However, an objective evaluation of the proposed algorithms is not possible due to the lack of biological ground truth information. To overcome this fundamental problem, the use of simulated microarray data for algorithm validation has been proposed.ResultsWe present a microarray simulation model which can be used to validate different kinds of data analysis algorithms. The proposed model is unique in the sense that it includes all the steps that affect the quality of real microarray data. These steps include the simulation of biological ground truth data, applying biological and measurement technology specific error models, and finally simulating the microarray slide manufacturing and hybridization. After all these steps are taken into account, the simulated data has realistic biological and statistical characteristics. The applicability of the proposed model is demonstrated by several examples.ConclusionThe proposed microarray simulation model is modular and can be used in different kinds of applications. It includes several error models that have been proposed earlier and it can be used with different types of input data. The model can be used to simulate both spotted two-channel and oligonucleotide based single-channel microarrays. All this makes the model a valuable tool for example in validation of data analysis algorithms.
| S-EPMC1574357 | biostudies-literature
Project description:The Hausa language read-speech dataset was created by recording native Hausa speakers. The recording took place at Nile university of Nigeria audio studio and radio broadcasting studio. The recorded dataset was segmented into unigram and bigram. The Hausa speech dataset contain 47hr of recorded audio speech. The dataset can be used for automatic speech recognition, speech synthesis, Text-to-Speech and speech-to-text application.
| S-EPMC8792404 | biostudies-literature
Project description:In the past few decades, deep learning algorithms have become more prevalent for signal detection and classification. To design machine learning algorithms, however, an adequate dataset is required. Motivated by the existence of several open-source camera-based hand gesture datasets, this descriptor presents UWB-Gestures, the first public dataset of twelve dynamic hand gestures acquired with ultra-wideband (UWB) impulse radars. The dataset contains a total of 9,600 samples gathered from eight different human volunteers. UWB-Gestures eliminates the need to employ UWB radar hardware to train and test the algorithm. Additionally, the dataset can provide a competitive environment for the research community to compare the accuracy of different hand gesture recognition (HGR) algorithms, enabling the provision of reproducible research results in the field of HGR through UWB radars. Three radars were placed at three different locations to acquire the data, and the respective data were saved independently for flexibility.
| S-EPMC8041886 | biostudies-literature