A fast and flexible implementation of parallel model combination

Abstract

In previous papers the use of Parallel Model Combination (PMC) for noise robustness has been described. Various fast implementations have been proposed, though to date in order to compensate all the parameters of a system it has been necessary to perform Gaussian integration. This pa per introduces an alternative method that can compensate all the parameters of the recognition system, whilst reduc ing the computational load of this task. Furthermore, the technique offers an additional degree of flexibility, as it al lows the number of components to be chosen and optimised using standard iterative techniques. The new technique is referred to as Data-driven PMC (DPMC). It is evaluated on the Resource Management database, with noise artifi cially added from the NOISEX-92 database. The perfor mance of DPMC is found to be comparable to PMC , at a far lower computational cost. In complex noise environ ments, by more accurately modelling the noise source, us ing multiple components, and then reducing the number of components to the original nnmber a slight improvement in performance is obtained.1.

Dereverberation and Denoising Techniques for ASR Applications

“…This approach attempts to estimate a noisy speech model from two other models: a previously trained one, based on clean speech, and a noise model, obtained by an on-line estimate from noise segments (Gales & Young, 1995). Promising adaptation results can be achieved by using a small amount of data, whereas the main drawback of the PMC approach is a large computational burden.…”

Auditory processing-based features for improving speech recognition in adverse acoustic conditions

“…Other variants include http://asmp.eurasipjournals.com/content/2014/1/21 mean-variance normalization (MVN), cepstral mean subtraction and variance normalization (CMSVN) and relative spectral (RASTA) filtering [2,6]. Model adaptation approaches modify the acoustic model parameters' match with the observed speech features [4,7].…”

Realce E Reconhecimento De Voz Contínua Em Ambientes Adversos

“…Como foi exposto na seção 2.3, a interação entre voz e ruído é expressa mais naturalmente no domínio do banco de filtros, assim o modelo HMM final é transformado ao domínio cepstral mediante a aplicação da transformada discreta do coseno DCT. Dessa técnica inúmeras variações têm sido apresentadas nas últimas décadas [166] [167][168] [169].…”