Combining Temporal and Spectral Features in HMM-Based Drum Transcription
نویسندگان
چکیده
To date several methods for transcribing drums from polyphonic music have been published. Majority of the features used in the transcription systems are “spectral”: parameterising some property of the signal spectrum in a relatively short time frames. It has been shown that utilising narrow-band features describing long-term temporal evolution in conjunction with the more traditional features can improve the overall performance in speech recognition. We investigate similar utilisation of temporal features in addition to the HMM baseline. The effect of the proposed extension is evaluated with simulations on acoustic data, and the results suggest that temporal features do improve the result slightly. Demonstrational signals of the transcription results are available at http://www.cs.tut.fi/sgn/arg/paulus/demo/. 1
منابع مشابه
Spectral-spatial classification of hyperspectral images by combining hierarchical and marker-based Minimum Spanning Forest algorithms
Many researches have demonstrated that the spatial information can play an important role in the classification of hyperspectral imagery. This study proposes a modified spectral–spatial classification approach for improving the spectral–spatial classification of hyperspectral images. In the proposed method ten spatial/texture features, using mean, standard deviation, contrast, homogeneity, corr...
متن کاملPhoneme Classification Using Temporal Tracking of Speech Clusters in Spectro-temporal Domain
This article presents a new feature extraction technique based on the temporal tracking of clusters in spectro-temporal features space. In the proposed method, auditory cortical outputs were clustered. The attributes of speech clusters were extracted as secondary features. However, the shape and position of speech clusters change during the time. The clusters temporally tracked and temporal tra...
متن کاملشبکه عصبی پیچشی با پنجرههای قابل تطبیق برای بازشناسی گفتار
Although, speech recognition systems are widely used and their accuracies are continuously increased, there is a considerable performance gap between their accuracies and human recognition ability. This is partially due to high speaker variations in speech signal. Deep neural networks are among the best tools for acoustic modeling. Recently, using hybrid deep neural network and hidden Markov mo...
متن کاملDrum Transcription from Polyphonic Music with Instrument-wise Hidden Markov Models
This paper describes a system for automatic transcription of drum instruments from polyphonic music signals. For each target drum instrument, a hidden Markov model (HMM) is created to describe the sound characteristics when the instrument is played. Also, a background model with only one state is created for each instrument to describe the sound when the target instrument is not played. The sig...
متن کاملAutomated Detection of Multiple Sclerosis Lesions Using Texture-based Features and a Hybrid Classifier
Background: Multiple Sclerosis (MS) is the most frequent non-traumatic neurological disease capable of causing disability in young adults. Detection of MS lesions with magnetic resonance imaging (MRI) is the most common technique. However, manual interpretation of vast amounts of data is often tedious and error-prone. Furthermore, changes in lesions are often subtle and extremely unrepresentati...
متن کامل