This book is the second edition of the handbook that was first published in 1999. It has 22 chapters in five parts: experimental phonetics, biological perspectives, modeling speech production and perception, linguistic phonetics, and speech technology.

There are four chapters in Part I, on experimental phonetics. In Chapter 1, Maureen Stone provides an overview of techniques for investigating articulation, including x-rays and MRI scans. In Chapter 2, Christine Shadle describes ways of measuring and modeling the aerodynamics of speech. In Chapter 3, Jonathan Harrington discusses the acoustics of vowels, obstruents and nasals. And in Chapter 4, Hajime Hirose deals with investigations into the structure and the operation of the larynx.

Part II, on biological perspectives, has three chapters. In Chapter 5, Janet Mackenzie Beck reviews the changes that take place in the vocal apparatus of men and women in the course of their lives, as well as the effects of pathological changes to the larynx, vocal tract and teeth. In Chapter 6, Hermann Ackermann & Wolfram Ziegler provide a summary of current knowledge about human brain mechanisms, comparing them with other primates, and they also review the effects of pathological damage to the brain and consider the findings of modern imaging techniques. And in Chapter 7, Anne Smith summarises recent research into changes in the movement of the lips and the jaw as well as neural mechanisms as children grow older.

Part III, on modeling speech production and perception, has seven chapters, making it the longest of the book. In Chapter 8, Barbara Davis offers a brief overview of models of speech acquisition by infants. In Chapter 9, Edda Farnetani & Daniel Recasens discuss connected speech processes, including whether there is any difference between assimilation (in which one sound affects a neighbouring sound) and coarticulation (which involves overlapping articulatory gestures). In Chapter 10, Anders Löfqvist considers models of speech production, especially the knowledge we can gain from experiments involving perturbations to speech movements. In Chapter 11, Christer Gobl & Ailbhe Ní Chasaide analyse voice source variation, describing methods of estimating the glottal flow, the use of labels such as breathy and creaky to describe voice quality, and the contribution of voice quality to the message. In Chapter 12, Kenneth Stevens & Helen Hanson present an overview of the quantal theory of speech production. In Chapter 13, Brian Moore gives a summary of auditory processing, particularly the findings of research involving threshold detection in the human auditory system. And in Chapter 14, James McQueen & Anne Cutler consider models of cognitive processes in speech perception, in particular comparing the TRACE and Merge models.

There are five chapters in Part IV, on linguistic phonetics. In Chapter 15, Janet Fletcher provides a substantial overview of issues involving the timing of speech, including the duration of consonants, vowels, syllables and pauses as well as the nature and measurement of rhythm. In Chapter 16, Mary Beckman & Jennifer Venditti discuss the representation of tone, and whether there is any basis for the three-way classification of tone languages, pitch-accent languages and intonation languages. In Chapter 17, John Ohala argues that it makes no sense to separate phonology from phonetics. In Chapter 18, John Esling gives an overview of the history and current status of the IPA symbols and the IPA chart. And in Chapter 19, Paul Foulkes, James Scobbie & Dominic Watt review the methods adopted by sociophonetic research and the contribution it can provide in understanding variation in pronunciation.

The final part of the book, Part V, on speech technology, contains just three chapters. In Chapter 20, Daniel Ellis gives a tutorial introduction to signal processing. In Chapter 21, Rolf Carlson & Björn Granström discuss the historical background and current state of speech synthesis. And in the last chapter, Chapter 22, Steve Renals & Simon King summarise work on automatic speech recognition, particularly algorithms based on Hidden Markov Models.

One issue with these various chapters is whether they should offer a detailed, densely packed state-of-the-art snapshot of the topic, or instead they should be more of a tutorial that could be understood by someone with little background knowledge, and there is quite a contrast in styles between, for example, Chapters 6 and 7. In Chapter 6, Ackermann & Ziegler offer a summary packed with specialist terminology about the brain, but much of it will be rather opaque to non-experts. For example, ‘the size of the hypoglossal nucleus was found enlarged as compared to the nonhuman haplorhine regression line, using the medulla oblongata as an independent variable’ (p. 213). I am afraid I am lost here. In contrast, in Chapter 7, everything that is explained by Smith is splendidly clear, even to someone like me, who has no background in research on neural networks and orofacial movements. But maybe experts in the two fields would regard the material in Chapter 6 as an exceptionally valuable, information-packed overview of the results of current research, while some people might find the material in Chapter 7 insufficiently detailed.

Other chapters with some densely packed material are Chapters 3 and 11. For example, in Chapter 3, Harrington provides some quite intricate analysis of shifts associated with the locus for F2 for the different plosives and vowels, and one would imagine that those not familiar with acoustic measurements might find some of this material a bit overwhelming. And in Chapter 11, Gobl & Ní Chasaide analyse measurements associated with voice source variation in admirable depth, including reference to EE (excitation strength), RA (dynamic leakage), OQ (open quotient), FG (glottal frequency), RK (glottal symmetry), AH (aspiration noise), FA (cutoff frequency), and various others parameters. While those with some background in these measurements are likely to find the coverage very thorough and exceptionally authoritative, it is hard for someone not familiar with all these parameters to keep track of them, particularly as the origin of some of the letters (such as RA and RK) remains obscure. Perhaps a list of abbreviations at the start of the chapter would have helped.

In contrast, Chapters 12 and 20 are targeted at people with no background in the subjects. For example, in Chapter 12, Stevens & Hanson explain with great clarity how a zero associated with a subglottal resonance can cause acoustic discontinuities at about 1400 Hz for the second formant, and this helps to explain the quantal basis for the feature [back] for vowels. And in Chapter 20, Ellis starts from fundamentals, including how the motion of a swing is built up by small, well-timed movements of the child's body in order to explain resonance, before he goes on to demonstrate how complex waveforms can be regarded as consisting of the summation of a series of sinusoidal waves. But perhaps readers with substantial expertise in acoustic theory and signal processing will find these two chapters of little value.

Another contrast between the various chapters is length. The longest is Chapter 15, which has 57 pages of text followed by 23 pages of references. In contrast, Chapter 8 consists of just twelve pages of text and five pages of references. It is not clear why Fletcher's description of timing and rhythm (Chapter 15) merits nearly five times the length of Davis's material on speech acquisition (Chapter 8). Of course, length is not always an arbiter of quality; but in this case the treatment of timing and rhythm seems very thorough, while that on speech acquisition appears somewhat cursory, listing various approaches without going into much detail.

Given that some of the material can be quite complex and densely packed, it is not helped when a few of the figures are flawed. For example, in Chapter 3, in Figure 3.13 (p. 109), the final spectrum is labelled [Δ] instead of the intended [ʃ], and in Figure 3.16 (p. 115), the left-hand spectrogram showing an extract of Warlpiri speech is labelled as [aɲam] while the caption below claims that it shows [aɲampu]. Then in Chapter 16, Figure 16.3 (p. 612) the text for fàng zài nèi-ge spoken in Beijing dialect is all merged together, making it very difficult to read. In Chapter 18, Figure 18.4 (p. 686) includes labels such as ‘FSryngeSl%voUSl%trSUt’, which presumably should be something else; and when discussing the positioning of the vowel quadrilateral in relation to the consonant table on the IPA chart, on page 697, it is stated that ‘the right edge should lie beneath the Velar column, as in the 1926 chart’, while in Figure 18.1 (p. 682), showing the 1926 chart, the right edge of the vowel quadrilateral is actually under the Uvular column, not the Velar column, which means that Figure 18.1 is incorrect (International Phonetic Association 1949: 10). In Chapter 21, none of the data listed in Figure 21.7 (p. 795) make any sense, with for example the final line suggesting that ‘iutomttiwEprowedures’ is one of the trends in speech synthesis development. I have no idea what this is supposed to be. Finally, in Chapter 22, the finite state language model depicted in Figure 22.5 (p. 823) seems to offer the chance to buy ‘thaee’ tickets to ‘(ondon’ or ‘(eeds’. Flaws such as these are a pity, as most of the material (including that in Chapters 3, 16, 18, 21 and 22) is highly informative, and all of it is written by acknowledged experts in the various fields, but flaws in these figures not only make the material harder to understand but they also lead to questions about the reliability of some of the information. For example, we find the Beijing dialect fàng zài nèi ge from Figure 16.3 transcribed as [fã.ða ne.ɣə] on page 613, and it remains unclear if zài is really pronounced as [ða] in Beijing dialect, especially as the second occurrence of this phrase on the same line is transcribed as [fã ne.ɣə], with the zài simply omitted, and furthermore Lee & Zee (Reference Lee and Zee2003) make no mention of [ð] as a possible sound in Beijing dialect.

Although the overall length of the second edition of this book is almost the same as that of the first, there are four fewer chapters, so each chapter is on average a bit longer (or has more references). Some chapters have been removed, including those on experimental design and statistics by William Shearer, phonetic fieldwork by Peter Ladefoged, phonetic universals by Ian Maddieson, and forensic phonetics by Francis Nolan. In particular, it seems a pity that there is now almost no material on forensic phonetics (apart from two brief mentions in Chapter 19), as this is such an important area of phonetic analysis nowadays. At the same time, there are some welcome extra chapters, including those on orofacial movements by Anne Smith, tone and intonation by Beckman & Venditti, the timing of speech by Fletcher, and sociophonetics by Foulkes, Scobbie & Watt, and some chapters have been re-written by new contributors, including acoustic phonetics by Harrington (replacing the original chapter by Osama Fujimura & Donna Erickson) and speech acquisition by Davis (replacing the chapter by Peter MacNeilage). All of these additional and re-written chapters include welcome new material, ensuring this handbook represents an invaluable and up-to-date overview of the field of phonetics that will ensure it remains an essential reference work for researchers and students for many years to come, even if there are a few flaws in the presentation.