Michael Picheny

Speaker adaptation of neural network acoustic models using i-vectors

We propose to adapt deep neural network (DNN) acoustic models to a target speaker by supplying speaker identity vectors (i-vectors) as input features to the network in parallel with the regular acoustic features for ASR. For both training and test, the i-vector for a given speaker is concatenated to every frame belonging to that speaker and changes across different speakers. Experimental results on a Switchboard 300 hours corpus show that DNNs trained on speaker independent features and i-vectors achieve a 10% relative improvement in word error rate (WER) over networks trained on speaker independent features only. These networks are comparable in performance to DNNs trained on speaker-adapted features (with VTLN and FMLLR) with the advantage that only one decoding pass is needed. Furthermore, networks trained on speaker-adapted features and i-vectors achieve a 5-6% relative improvement in WER after hessian-free sequence training over networks trained on speaker-adapted features only.

Speech recognition using noise-adaptive prototypes

A probabilistic mixture mode is described for a frame (the short term spectrum) of speech to be used in speech recognition. Each component of the mixture is regarded as a prototype for the labeling phase of a hidden Markov model based speech recognition system. Since the ambient noise during recognition can differ from that present in the training data, the model is designed for convenient updating in changing noise. Based on the observation that the energy in a frequency band is at any fixed time dominated either by signal energy or by noise energy, the energy is modeled as the larger of the separate energies of signal and noise in the band. Statistical algorithms are given for training this as a hidden variables model. The hidden variables are the prototype identities and the separate signal and noise components. Speech recognition experiments that successfully utilize this model are described. >

Acoustic Markov models used in the Tangora speech recognition system

The Speech Recognition Group at IBM Research has developed a real-time, isolated-word speech recognizer called Tangora, which accepts natural English sentences drawn from a vocabulary of 20000 words. Despite its large vocabulary, the Tangora recognizer requires only about 20 minutes of speech from each new user for training purposes. The accuracy of the system and its ease of training are largely attributable to the use of hidden Markov models in its acoustic match component. An automatic technique for constructing Markov word models is described and results are included of experiments with speaker-dependent and speaker-independent models on several isolated-word recognition tasks.<<ETX>>

Deep Belief Networks using discriminative features for phone recognition

Deep Belief Networks (DBNs) are multi-layer generative models. They can be trained to model windows of coefficients extracted from speech and they discover multiple layers of features that capture the higher-order statistical structure of the data. These features can be used to initialize the hidden units of a feed-forward neural network that is then trained to predict the HMM state for the central frame of the window. Initializing with features that are good at generating speech makes the neural network perform much better than initializing with random weights. DBNs have already been used successfully for phone recognition with input coefficients that are MFCCs or filterbank outputs [1, 2]. In this paper, we demonstrate that they work even better when their inputs are speaker adaptive, discriminative features. On the standard TIMIT corpus, they give phone error rates of 19.6% using monophone HMMs and a bigram language model and 19.4% using monophone HMMs and a trigram language model.

Performance of the IBM large vocabulary continuous speech recognition system on the ARPA Wall Street Journal task

In this paper we discuss various experimental results using our continuous speech recognition system on the Wall Street Journal task. Experiments with different feature extraction methods, varying amounts and type of training data, and different vocabulary sizes are reported.

Decision trees for phonological rules in continuous speech

The authors present an automatic method for modeling phonological variation using decision trees. For each phone they construct a decision tree that specifies the acoustic realization of the phone as a function of the context in which it appears. Several-thousand sentences from a natural language corpus spoken by several speakers are used to construct these decision trees. Experimental results on a 5000-word vocabulary natural language speech recognition task are presented.<<ETX>>

Automatic recognition of spontaneous speech for access to multilingual oral history archives

Much is known about the design of automated systems to search broadcast news, but it has only recently become possible to apply similar techniques to large collections of spontaneous speech. This paper presents initial results from experiments with speech recognition, topic segmentation, topic categorization, and named entity detection using a large collection of recorded oral histories. The work leverages a massive manual annotation effort on 10 000 h of spontaneous speech to evaluate the degree to which automatic speech recognition (ASR)-based segmentation and categorization techniques can be adapted to approximate decisions made by human annotators. ASR word error rates near 40% were achieved for both English and Czech for heavily accented, emotional and elderly spontaneous speech based on 65-84 h of transcribed speech. Topical segmentation based on shifts in the recognized English vocabulary resulted in 80% agreement with manually annotated boundary positions at a 0.35 false alarm rate. Categorization was considerably more challenging, with a nearest-neighbor technique yielding F=0.3. This is less than half the value obtained by the same technique on a standard newswire categorization benchmark, but replication on human-transcribed interviews showed that ASR errors explain little of that difference. The paper concludes with a description of how these capabilities could be used together to search large collections of recorded oral histories.

The metamorphic algorithm: a speaker mapping approach to data augmentation

Large vocabulary speaker-dependent speech recognition systems adjust to the acoustic peculiarities of each new speaker based on some enrolment data provided by this speaker. As the amount of data required increases with the sophistication of the underlying acoustic models, the enrolment may get lengthy. To streamline it, it is therefore desirable to make use of previously acquired speech data. The authors describe a data augmentation strategy based on a piecewise linear mapping between the feature space of a new speaker and that of a reference speaker. This speaker-normalizing mapping is used to transform the previously acquired data of the reference speaker onto the space of the new speaker. The performance of the resulting procedure, dubbed the metamorphic algorithm, is illustrated on an isolated utterance speech recognition task with a vocabulary of 20000 words. Results show that the metamorphic algorithm can substantially reduce the word error rate when only a limited amount of enrolment data is available. Alternatively, it leads to a level of performance comparable to that obtained when a much greater amount of enrolment data is required from the new speaker. In addition, it can also be used for tracking spectral evolution over time, thus providing a possible means for robust speaker self-adaptation. >

Multonic Markov word models for large vocabulary continuous speech recognition

A new class of hidden Markov models is proposed for the acoustic representation of words in an automatic speech recognition system. The models, built from combinations of acoustically based sub-word units called fenones, are derived automatically from one or more sample utterances of a word. Because they are more flexible than previously reported fenone-based word models, they lead to an improved capability of modeling variations in pronunciation. They are therefore particularly useful in the recognition of continuous speech. In addition, their construction is relatively simple, because it can be done using the well-known forward-backward algorithm for parameter estimation of hidden Markov models. Appropriate reestimation formulas are derived for this purpose. Experimental results obtained on a 5000-word vocabulary natural language continuous speech recognition task are presented to illustrate the enhanced power of discrimination of the new models. >

Speech recognizer having a speech coder for an acoustic match based on context-dependent speech-transition acoustic models

A speech coding apparatus compares the closeness of the feature value of a feature vector signal of an utterance to the parameter values of prototype vector signals to obtain prototype match scores for the feature vector signal and each prototype vector signal. The speech coding apparatus stores a plurality of speech transition models representing speech transitions. At least one speech transition is represented by a plurality of different models. Each speech transition model has a plurality of model outputs, each comprising a prototype match score for a prototype vector signal. Each model output has an output probability. A model match score for a first feature vector signal and each speech transition model comprises the output probability for at least one prototype match score for the first feature vector signal and a prototype vector signal. A speech transition match score for the first feature vector signal and each speech transition comprises the best model match score for the first feature vector signal and all speech transition models representing the speech transition. The identification value of each speech transition and the speech transition match score for the first feature vector signal and each speech transition are output as a coded utterance representation signal of the first feature vector signal.