Various technological, economic, and social developments may radically reshape engineering and computer science education in the next 20 to 30 years. Leaders in engineering education, such as IEEE, that play key roles must decide how they will innovate and support initiatives to improve engineering education. Decisions must be guided by both historical trends and potential future evolutions of engineering education. Since the future cannot be predicted, the organizations may be interested in methodologies to help make decisions. IEEE's Curriculum and Pedagogical Committee (CPC), which is charged with helping IEEE make solid decisions, engaged in a scenario planning exercise to help understand how various influences and trajectories could influence engineering education. Scenario planning methodology has been used by many organizations making long-term investments, to understand how future developments could be taken into account when making key decisions. The purpose of the workshop is to allow participants to experience some aspects of the scenario planning methodology and to compare their results with the results generated by the CPC.

Various technological, economic, and social developments may radically reshape engineering and computer science education in the next 20 to 30 years. Leaders in engineering education, such as IEEE, that play key roles must decide how they will innovate and support initiatives to improve engineering education. The IEEE's Curriculum and Pedagogical Committee (CPC), which is charged with helping IEEE forecast the future of engineering education and make decisions about IEEE's roles in preparing for and crafting that future, engaged in a scenario planning exercise to help understand how various influences and trajectories could influence engineering education. Scenario planning methodology has been used by many organizations making long-term investments. After considering influencers in the engineering education environment; important stakeholders and their roles; goals, constraints, and outcomes for key stakeholders; and potential uncertainties, the CPC identified two critical uncertainties over the next five years: (i) How will engineering programs innovate and adapt? (ii) What will be the values and competencies of engineering faculty? By considering two extremes for each of these uncertainties, the CPC crafted four scenarios, which are described in this paper, to characterize potential future trends and how different scenarios will influence the evolution of engineering education.

For query-by-example spoken term detection (QbE-STD), generation of phone posteriorgrams requires labelled data which would be difficult for languages with low resources. One solution is to build models from rich resource languages and use them in the low resource scenario. However, phone classes are not language universal and alternate representation such as articulatory classes is explored. In this paper, we use articulatory information and their derivatives such as bottle-neck (BN) features (also referred to as articulatory BN features) for QbE-STD. We obtain Gaussian posteriorgrams of articulatory BN features in tandem with the acoustic parameters such as frequency domain linear prediction cepstral coefficients to perform the search. We compare the search performance of articulatory and phone BN features and show that articulatory BN features are a better representation. We also provide experimental results to show that low amounts (30 mins) of training data could be used to derive articulatory BN features.

The task of query-by-example spoken term detection (QbE-STD) is to find a spoken query within spoken audio data. Current state-of-the-art techniques assume zero prior knowledge about the language of the audio data, and thus explore dynamic time warping (DTW) based techniques for the QbE-STD task. In this paper, we use a variant of DTW based algorithm referred to as non-segmental DTW (NS-DTW), with a computational upper bound of O (mn) and analyze the performance of QbE-STD with Gaussian posteriorgrams obtained from spectral and temporal features of the speech signal. The results show that frequency domain linear prediction cepstral coefficients, which capture the temporal dynamics of the speech signal, can be used as an alternative to traditional spectral parameters such as linear prediction cepstral coefficients, perceptual linear prediction cepstral coefficients and Mel-frequency cepstral coefficients. We also introduce another variant of NS-DTW called fast NS-DTW (FNS-DTW) which uses reduced feature vectors for search. With a reduction factor of α ∈ ℕ, we show that the computational upper bound for FNS-DTW is O(mn/(α2)) which is faster than NS-DTW.

This paper describes some of the results from the project entitled “New Parameterization for Emotional Speech Synthesis” held at the Summer 2011 JHU CLSP workshop. We describe experiments on how to use articulatory features as a meaningful intermediate representation for speech synthesis. This parameterization not only allows us to reproduce natural sounding speech but also allows us to generate stylistically varying speech.

In this paper, we describe the “Spoken Web Search” Task, which was held as part of the 2011 MediaEval benchmark campaign. The purpose of this task was to perform audio search with audio input in four languages, with very few resources being available in each language. The data was taken from “spoken web” material collected over mobile phone connections by IBM India. We present results from several independent systems, developed by five teams and using different approaches, compare them, and provide analysis and directions for future research.

In this paper, we propose modeling a noisy-channel for the task of voice conversion (VC). We have used the artificial neural networks (ANN) to capture speaker-specific characteristics of a target speaker which avoid the need for any training utterance from a source speaker. We use articulatory features (AFs) as a canonical form or speaker-independent representation of a speech signal. Our studies show that AFs contain a significant amount of speaker information in their trajectories. Suitable techniques are proposed to normalize the speaker-specific information in AF trajectories and the resultant AFs are used in voice conversion. The results of voice conversion evaluated using objective and subjective measures confirm that AFs can be used as a canonical form in nosiy-channel to capture speakerspecific characteristics of a target speaker.

This paper discusses the efforts in collecting speech databases for Indian languages – Bengali, Hindi, Kannada, Malayalam, Marathi, Tamil and Telugu. We discuss relevant design considerations in collecting these databases, and demonstrate their usage in speech synthesis. By releasing these speech databases in the public domain without any restrictions for non commercial and commercial purposes, we hope to promote research and developmental activities in building speech synthesis systems in Indian languages. Index Terms:speech databases, speech synthesis, Indian languages

Zero frequency filtering is a technique used in the characterization and analysis of glottal activity from speech signals. The filter design originally proposed has an infinite impulse response (IIR) filter followed by two successive finite impulse response (FIR) filters. In this paper, the process of its computation is analyzed and a simplified FIR implementation is proposed by employing the inherent pole-zero cancellation involved in the process. Theoretical proofs are derived in both, frequency and time domains. We show that the theoretically derived FIR filter is a convolution of two filters, whose impulse responses are triangular shaped. The advantage of the proposed FIR filter lies in reduction of computational requirements for zero frequency filtering which include - 1) use of single-precision floating point and 2) stability of the filter.

Unit selection synthesis inventories have coverage issues, which lead to missing syllable or diphone units. In the conventional back-off strategy of substituting the missing unit with approximate unit(s), the rules for approximate matching are hard to derive. In this paper we propose a back-off strategy for Telugu TTS systems emulating native speaker intuition. It uses reduced vowel insertion in complex consonant clusters to replace missing units. The inserted vowel identity is determined using a rule-set adapted from L2 (second language) acquisition research in Telugu, reducing the effort required in preparing the rule-set. Subjective evaluations show that the proposed back-off method performs better than the conventional methods.

High accuracy speech segmentation methods invariably depend on manually labelled data. However under-resourced languages do not have annotated speech corpora required for training these segmentors. In this paper we propose a boundary refinement technique which uses knowledge of phone-class specific subband energy events, in place of manual labels, to guide the refinement process. The use of this knowledge enables proper placement of boundaries in regions with multiple spectral discontinuities in close proximity. It also helps in the correction of large alignment errors. The proposed refinement technique provides boundaries with an accuracy of 82% within 20ms of actual boundary. Combining the proposed technique with iterative isolated HMM training technique boosts the accuracy to 89%, without the use of any manually labelled data.

One of the issues in using audio books for building a synthetic voice is the segmentation of large speech files. The use of the Viterbi algorithm to obtain phone boundaries on large audio files fails primarily because of huge memory requirements. Earlier works have attempted to resolve this problem by using large vocabulary speech recognition system employing restricted dictionary and language model. In this paper, we propose suitable modifications to the Viterbi algorithm and demonstrate its usefulness for segmentation of large speech files in audio books. The utterances obtained from large speech files in audio books are used to build synthetic voices. We show that synthetic voices built from audio books in the public domain have Mel-cepstral distortion scores in the range of 4-7, which is similar to voices built from studio quality recordings such as CMU ARCTIC.

In this work, we compare different approaches for speech segmentation, of which some are constrained and the remaining are unconstrained by phone transcript. A high accuracy speech segmentation can be obtained by approaches constrained by phone transcript such as HMM forced-alignment when exact phone transcript is known. But such approaches have to adjust with canonical phone transcript ,a sexact phone transcript is tough to obtain. Our experiments on TIMIT corpus demonstrate that ANN and HMM phone-loop based unconstrained approaches, perform better than HMM forced-alignment based approach constrained by canonical phone transcript. Finally a detailed error analysis of these approaches is reported. Index Terms: HMM, Group-Delay, ANN, Speech Segmentation.

In this paper we propose prominence based features for ranking speech segments for automatic speech summmarization. Standard speech summarization systems depend on ASR transcripts or/and gold standard human reference summaries which limits application of such systems. The proposed method uses prominence values of syllables in a speech segment to rank the segment for summarization. The proposed method does not depend on ASR transcripts or gold standard human summaries. Evaluation results showed that summaries generated by the proposed method can generate summaries as good as the summaries generated using tf*idf scores and supervised system trained on gold standard summaries. Experiments are carried out on two types of speech corpora one containing read style news speech and the other spontaneous telephone conversations.

Analysis of human reference summaries of broadcast news showed that humans give preference to anchor speaker segments while constructing a summary. Therefore, we exploit the role of anchor speaker in a news show by tracking his/her speech to construct indicative/informative extractive audio summaries. Speaker tracking is done by Bayesian information criterion (BIC) technique. The proposed technique does not require Automatic Speech Recognition (ASR) transcripts or human reference summaries for training. The objective evaluation by ROUGE showed that summaries generated by the proposed technique are as good as summaries generated by a baseline text summarization system taking manual transcripts as input and summaries generated by a supervised speech summarization system trained using human summaries. The subjective evaluation of audio summaries by humans showed that they prefer summaries generated by proposed technique to summaries generated by supervised speech summarization system.

In this paper, we use artificial neural networks (ANNs) for voice conversion and exploit the mapping abilities of an ANN model to perform mapping of spectral features of a source speaker to that of a target speaker. A comparative study of voice conversion using an ANN model and the state-of-the-art Gaussian mixture model (GMM) is conducted. The results of voice conversion, evaluated using subjective and objective measures, confirm that an ANN-based VC system performs as good as that of a GMM-based VC system, and the quality of the transformed speech is intelligible and possesses the characteristics of a target speaker. In this paper, we also address the issue of dependency of voice conversion techniques on parallel data between the source and the target speakers. While there have been efforts to use nonparallel data and speaker adaptation techniques, it is important to investigate techniques which capture speaker-specific characteristics of a target speaker, and avoid any need for source speaker's data either for training or for adaptation. In this paper, we propose a voice conversion approach using an ANN model to capture speaker-specific characteristics of a target speaker and demonstrate that such a voice conversion approach can perform monolingual as well as cross-lingual voice conversion of an arbitrary source speaker.

In this paper, we propose to use artificial neural networks (ANN) for voice conversion. We have exploited the mapping abilities of ANN to perform mapping of spectral features of a source speaker to that of a target speaker. A comparative study of voice conversion using ANN and the state-of-the-art Gaussian mixture model (GMM) is conducted. The results of voice conversion evaluated using subjective and objective measures confirm that ANNs perform better transformation than GMMs and the quality of the transformed speech is intelligible and has the characteristics of the target speaker.

In this paper we present our argument that context information could be used in early stages i.e., during the definition of mapping of the words into sequence of graphemes. We show that the early tagged contextual graphemes play a significant role in improving the performance of grapheme based speech synthesis and speech recognition systems.

Pattern classification is an important task in speech recognition and speaker verification. Given the feature vectors of an input the goal is to capture the characteristics of these features unique to each class. This paper deals with exploring Auto Associative Neural Network (AANN) models for the task of speaker verification and speech recognition. We show that AANN models produce comparable performance with that of GMM based speaker verification and speech recognition.

In this paper, we are investigating the unit size: syllable, half-phone and quarter-phone to be used for speech synthesis in multi-lingual screen reader in phonetic languages such as Telugu and non-phonetic language English. Perceptual studies show that syllable-level unit performs better for Telugu and half-phone units perform better for English. While syllable based synthesizers produce better sounding speech, the coverage of all syllables is a non-trivial issue. We address the issue of coverage of syllables through approximate matching of syllable and show that such approximation produces intelligible and better quality speech than diphone units. In this paper, we also propose a hybrid synthesizer within the framework of unit selection and also show that the hybrid synthesizer built from pruned database performs as well as hybrid synthesizer built from unpruned database. Index Terms: speech synthesis, unit selection, unit size, database pruning, and hybrid speech synthesis.

Letter-to-sound (LTS) rules play a vital role in building a speech synthesis system. In this paper, we apply various Machine Learning approaches like Classication and Regression Trees (CART), Decision Forest, forest of Articial Neural Network (ANN) and Auto Associative Neural Networks (AANN) for LTS rules. We used these techniques mainly for Schwa dele-tion in Hindi. We empirically show that the LTS using Decision Forest and Forest of ANNs outperforms the previous CART and normal ANN approaches respectively, and the non discrimina-tive learning technique of AANN could not capture the LTS rules as efciently as discriminative techniques. We explore use of syllabic features, namely, syllabic structure, onset of the syllable, number of syllables and place of Schwa along with pri-mary contextual features. The results showed that use of these features leads to good performance. The Decision Forest and forest of ANNs approaches yielded phone accuracy of 92.86% and 93.18 % respectively using the newly incorporated features for Hindi LTS.

Indian languages are syllabic in nature where many syllables are found common across its languages. This motivates us to build a global syllable set by combining multiple language syllables to build a synthesizer which can borrow units from a different language when the required syllable is not found. Such synthesizer make use of speech database in different languages spoken by different speakers, whose output is likely to pick units from multiple languages and hence the synthesized utterance contains units spoken by multiple speakers which would annoy the user. We intend to use a cross lingual Voice Conversion framework using Artificial Neural Networks (ANN) to transform such an utterance to a single target speaker.

In this paper we propose a technique for a syllable based speech synthesis system. While syllable based synthesizers produce better sounding speech than diphone and phone, the coverage of all syllables is a non-trivial issue. We address the issue of coverage of syllables through approximating the syllable when the required syllable is not found. To verify our hypothesis, we conducted perceptual studies on manually modified sentences and found that our assumption is valid. Similar approaches have been used in speech synthesis and it shows that such approximation produces intelligible and better quality speech than diphone units.

Linguistic knowledge plays a major role in building high quality speech systems such as speech synthesizer and speech recognizer systems. Quality of such speech systems greatly depends on the availability of linguistic knowledge such as pronunciation dictionary, stress pattern, syllable structure and so on. However, it is very difficult to obtain these linguistic knowledge in the required format for most of the languages. In this paper an attempt has been made to design and develop tools for extraction of the core linguistic information (letter to sound rule) automatically with the aid of acoustic data to build improved quality speech systems. Acoustic evidences will be exploited with minimal language knowledge to deploy the vital linguistic resources for such systems. The resulting letter to sound (LTS) rule is tested on different test data on Amharic, Hindi and Tamil. The performance of the letter-to-sound rule is reported in this paper.

Large multi paragraph speech databases encapsulate prosodic and contextual information beyond the sentence level which could be exploited to build natural sounding voices. This paper discusses our efforts on automatic building of synthetic voices from large multi-paragraph speech databases. We show that the primary issue of segmentation of large speech file could be addressed with modifications to forced-alignment technique and that the proposed technique is independent of the duration of the audio file. We also discuss how this framework could be extended to build a large number of voices from public domain large multi-paragraph recordings. Index Terms: speech synthesis, large multi-paragraph speech databases, forced-alignment, public domain recordings

In this paper we address the issue of pronunciation modeling for conversational speech synthesis. We experiment with two different HMM topologies (fully connected state model and forward connected state model) for sub-phonetic modeling to capture the deletion and insertion of sub-phonetic states during speech production process. We show that the experimented HMM topologies have higher log likelihood than the traditional 5-state sequential model. We also study the first and second mentions of content words and their influence on the pronunciation variation. Finally we report phone recognition experiments using the modified HMM topologies

In this paper, we describe a prototype speaker identification system using auto-associative neural network (AANN) and formant features. Our experiments demonstrate that formants extracted from difference spectrum perform significantly better than formants extracted from normal spectrum for the task of speaker identification. We also demonstrate that formants from difference spectrum provide comparable speaker identification performance with that of features such as weighted linear predictive Cepstral coefficients and Mel-Frequency Cepstral coefficients. Finally, we combine the results of formant based system and linear predictive Cepstral coefficients based system to achieve 100 % identification performance. Index Terms: Formant extraction, difference spectrum, speaker identification, autoassociative neural network models

In the paper we present our experiments with automatic pruning of speech databases created by us for Unit Selection based speech synthesis systems. Several algorithms have been attempted and perceptually evaluated. An optimal size of speech database has been reached where lose of naturalness due to unit pruning is not perceptible. 1.

In this paper, we address the issue of choice of unit size in unit selection speech synthesis. We discuss the development of a Hindi speech synthesizer and our experiments with different choices of units: syllable, diphone, phone and half phone. Perceptual tests conducted to evaluate the quality of the synthesizers with different unit size indicate that the syllable synthesizer performs better than the phone, diphone and half phone synthesizers, and the half phone synthesizer performs better than diphone and phone synthesizers.

We study the effectiveness of the features extracted from the source and system components of the speech production process for the purpose of speaker recognition. The source and system components are derived using linear prediction (LP) analysis of short segments of speech. The source component is the LP residual derived from the signal, and the system component is a set of weighted linear prediction cepstral coefficients. The features are captured implicitly by a feedforward autoassociative neural network (AANN). Two separate speaker models are derived by training two AANN models using feature vectors corresponding to source and system components. A speaker recognition system for 20 speakers is built and tested using both the models to evaluate the performance of source and system features. The study demonstrates the complementary nature of the two components.

The characteristics of the telephone channel and handset have a significant effect on the performance of speaker verification systems. The channel/handset mismatch between the training and testing data degrades the performance of speaker verification systems. In this paper, we show that the autoassociative neural network (AANN) models can be used to minimize the effects of channel characteristics on the performance of a text-independent speaker verification system. This paper also compares two approaches to represent the background model for an AANN based speaker verification system.