This paper describes the application of the harmonic plus noise model (HNM) for concatenative text-to-speech (TTS) synthesis. In the context of HNM, speech signals are represented as a time-varying harmonic component plus a modulated noise component. The decomposition of a speech signal into these two components allows for more natural-sounding modifications of the signal (e.g., by using different and better adapted schemes to modify each component). The parametric representation of speech using HNM provides a straightforward way of smoothing discontinuities of acoustic units around concatenation points. Formal listening tests have shown that HNM provides high-quality speech synthesis while outperforming other models for synthesis (e.g., TD-PSOLA) in intelligibility, naturalness, and pleasantness.

/pdf/applying-the-harmonic-plus-noise-model-in-concatenative-4rwr01lka3.pdf

Applying the harmonic plus noise model in concatenative speech synthesis

Analytical background and techniques: discrete-time signals, systems and transforms analysis of discrete-time speech signals probability and random processes linear model and dynamic system model optimization methods and estimation theory statistical pattern recognition Fundamentals of speech science: phonetic process phonological process Computational phonology and phonetics: computational phonology computational models for speech production computational models for auditory speechprocessing Speech technology in selected areas: speech recognition speech enhancement speech synthesis

/pdf/speech-processing-a-dynamic-and-optimization-oriented-2ml845xvfe.pdf

Speech processing : a dynamic and optimization-oriented approach

The new AT&T TTS system for general U.S. English text is based on best‐choice components picked from the AT&T Flextalk TTS, the Festival System from the University of Edinburgh, and ATR’s CHATR system. From Flextalk, it employs text normalization, letter‐to‐sound, and (optionally) baseline prosody generation. Festival provides general software‐engineering infrastructure (modularity) for easy experimentation and competitive evaluation of different algorithms or modules. Finally, CHATR’s unit selection was modified to guarantee the intelligibility of a good n‐phone (n=2 would be diphone) synthesizer while improving significantly on perceived naturalness relative to Flextalk. Each decision made during the research and development phase of this system was based on formal subjective evaluations. For example, the best voice found in a test that compared TTS systems built from several speakers gave a 0.3‐point head start (on a 5‐point rating scale) in quality over the mean of all speakers. Similarly, using our Harmonic‐plus‐Noise speech representation gave us a 0.25‐point advantage over standard TD‐PSOLA. Finally, not performing prosodic modifications (other than some smoothing across concatenation points) on the units but using the system‐generated prosody as a target in unit selection, 0.4 points were gained on overall quality. In conclusion, the new system combines the best of rule‐based and data‐driven worlds in TTS technology to deliver on the long‐standing promise of truly natural‐sounding synthesis.

The AT & T NEXT-GEN TTS system

This paper examines how people communicate with computers using speech. Automatic speech recognition (ASR) transforms speech into text, while automatic speech synthesis [or text-to-speech (TTS)] performs the reverse task. ASR has been largely developed based on speech coding theory, while simulating certain spectral analyses performed by the ear. Typically, a Fourier transform is employed, but following the auditory Bark scale and simplifying the spectral representation with a decorrelation into cepstral coefficients. Current ASR provides good accuracy and performance on limited practical tasks, but exploits only the most rudimentary knowledge about human production and perception phenomena. The popular mathematical model called the hidden Markov model (HMM) is examined; first-order HMMs are efficient but ignore long-range correlations in actual speech. Common language models use a time window of three successive words in their syntactic-semantic analysis. Speech synthesis is the automatic generation of a speech waveform, typically from an input text. As with ASR, TTS starts from a database of information previously established by analysis of much training data, both speech and text. Previously analyzed speech is stored in small units in the database, for concatenation in the proper sequence at runtime. TTS systems first perform text processing, including "letter-to-sound" conversion, to generate the phonetic transcription. Intonation must be properly specified to approximate the naturalness of human speech. Modern synthesizers using large databases of stored spectral patterns or waveforms output highly intelligible synthetic speech, but naturalness remains to be improved.

Interacting with computers by voice: automatic speech recognition and synthesis

A method for speech synthesis includes receiving an input speech signal containing a set of speech segments, and estimating spectral envelopes of the input speech signal in a succession of time intervals during each of the speech segments. The spectral envelopes are integrated over a plurality of window functions in a frequency domain so as to determine elements of feature vectors corresponding to the speech segments. An output speech signal is reconstructed by concatenating the feature vectors corresponding to a sequence of the speech segments.

Feature-domain concatenative speech synthesis

In an effort to select a speech representation for our next generation concatenative text-to-speech synthesizer, the use of two candidates is investigated; TD-PSOLA and the harmonic plus noise model, HNM. A formal listening test has been conducted and the two candidates have been rated regarding intelligibility, naturalness and pleasantness. Ability for database compression and computational load is also discussed. The results show that HNM consistently outperforms TD-PSOLA in all the above features except for computational load. HNM allows for high-quality speech synthesis without smoothing problems at the segmental boundaries and without buzziness or other oddities observed with TD-PSOLA.

L. Garrison

Papers

TD-PSOLA versus harmonic plus noise model in diphone based speech synthesis