Audio-visual Synchronization Research Articles

Gamified learning experience based on digital media and the application of virtual teaching robots in dance teaching

The research aims to explore the application of gamified learning experience based on digital media and virtual teaching robots in dance teaching. A virtual teaching robot system is designed, which combines technologies such as 3D motion capture, pose calculation, loss function constraint, and similarity matching. Through methods such as bone orientation, digital media audio-visual synchronization, and analysis of robot motion effects, virtual teaching robots have been able to present dance movements. Integrating virtual teaching robots into the teaching process through gamified learning methods. By introducing elements such as tasks, challenges, and rewards into learning, it is possible to stimulate students’ interest and enthusiasm in learning, and improve learning outcomes. The application of digital media based virtual teaching robots in dance teaching has a good effect. The overall structure of the system is reasonable, and the application of virtual reality technology allows students to learn dance firsthand. Through the analysis of skill teaching effectiveness and comprehensive teaching effectiveness, it was found that students have better participation and learning outcomes when learning dance through virtual teaching robots. The implementation of this system can enhance students’ learning enthusiasm and effectiveness, providing a new way for innovation in dance teaching.

Relentless Melt: visual trends and aesthetics of contemporary visual music films in Hong Kong

In recent years, a growing trend of abstract and experimental animation has emerged in Hong Kong, with an increasing number of works produced over the last two decades, particularly since around 2015. This article examines how Hong Kong artists approach abstract animation, the predominant aesthetic trends, and how these trends relate to or diverge from Western traditions in visual music, following pioneers such as Hans Richter, Viking Eggeling, Oskar Fischinger, Harry Smith and Norman McLaren. Employing content analysis, as used by Emmanouil Kanellos (2018) and textual analysis, this study investigates 95 selected works from the Relentless Melt screening series curated by Max Hattler and organized by the Hong Kong-based society Relentless Melt. By categorizing the data under two separate systems – the ‘What System’ and the ‘Where System’, as suggested by Margaret Livingstone in Vision and Art: The Biology of Seeing (2002) – the analysis encompasses aspects related to the perception or recognition of objects, faces, or other properties under the ‘What System’, and the perception of motion, depth, spatial organization and figure/ground segregation under the ‘Where System’. Unlike pioneering visual music artists who largely abandoned mimetic constructions in their aesthetics, some Hong Kong artists employ various types of representational abstraction. In addition, many do not shy away from expressing concepts or narratives in their hybrid visual music works while retaining fundamental elements, such as audio-visual synchronization. This project represents an initial step towards a more systematic understanding of Hong Kong’s visual music practices, aiming to promote plurality over homogeneity by critically examining the multiple facets of artistic practice in Hong Kong. It is hoped that this study will contribute to international scholarly discourse and establish a trajectory for further exploration.

AS-Net: active speaker detection using deep audio-visual attention

Active Speaker Detection (ASD) aims at identifying the active speaker among multiple speakers in a video scene. Previous ASD models often seek audio and visual features from long video clips with a complex 3D Convolutional Neural Network (CNN) architecture. However, models based on 3D CNNs can generate discriminative spatial-temporal features, but this comes at the expense of computational complexity, and they frequently face challenges in detecting active speakers in short video clips. This work proposes the Active Speaker Network (AS-Net) model, a simple yet effective ASD method tailored for detecting active speakers in relatively short video clips without relying on 3D CNNs. Instead, it incorporates the Temporal Shift Module (TSM) into 2D CNNs, facilitating the extraction of dense temporal visual features without the need for additional computations. Moreover, self-attention and cross-attention schemes are introduced to enhance long-term temporal audio-visual synchronization, thereby improving ASD performance. Experimental results demonstrate that AS-Net outperforms state-of-the-art 2D CNN-based methods on the AVA-ActiveSpeaker dataset and remains competitive with the methods utilizing more complex architectures.

Analysis the structural complexity of “Time” in Hans Zimmer’s score for ”Inception”

This study explores the intricate structure of "Time", a piece in Hans Zimmer's. The study aims to analyze the complex layers of "Time" and its influence on the overall impression of the picture. The primary study question investigates how the structural components of the composition, such as its recurring motifs, slow progression, and emotional climaxes, correspond to and amplify the movie's examination of time and reality. The process entails a qualitative analysis utilizing a literature review and an observational assessment of the film's audio-visual synchronization. By analyzing scholarly publications and conducting a literature study, the research reveals how Zimmer's minimalist approach and his inventive utilization of electronic sounds produce a distinctive auditory encounter that reflects the film's themes. Examining pivotal scenes that prominently display the concept of "Time" allows us to understand the intentional decisions made in the composition and execution of the music. The research findings indicate that the intricate composition of "Time" greatly amplifies the viewer's sense of being fully engaged in the film. The changing dynamics and recurring patterns of the piece not only mirror the emotional progress of the characters but also act as a guiding force for the viewer, helping them navigate through the complex storyline of the film. The emotional trajectory depicted in "Time"—evolving from a state of tension and confusion to one of resolution and hope—mirrors the thematic trajectory of the film, emphasizing its crucial function as an integral element in the narrative process. This analysis illustrates that the popularity of "Inception" is not exclusively attributed to its visual and narrative originality but is equally ascribed to its refined utilization of music. "Time" demonstrates how film music may significantly impact the audience's emotional and intellectual involvement, solidifying its position as a significant piece in modern cinema.

The influence of temporal (a)synchrony and linguistic complexity on audiovisual speech perception

Audiovisual (AV) cues often lead to improved speech understanding, especially when the auditory input is impoverished (e.g., in background noise, for people with hearing loss). This AV improvement occurs when both the auditory and visual information can be integrated together effectively (e.g., through temporal synchrony). However, whether the complexity of speech signals influences this integrated percept and how this contributes to speech understanding is still unknown. This study investigates whether individual differences in perception of temporal synchrony of the auditory and visual speech signal explain individual variability in AV speech recognition, and whether this differs across different levels of linguistic processing. By implementing five different levels of linguistic complexity (i.e., varying phoneme-viseme connections, vocabulary knowledge and/or linguistic context) and introducing temporal asynchrony in a remote task with English speaking adults we ask the following questions: (1) Is temporal synchrony needed to observe AV speech perception benefits? (2) Does this vary by linguistic level? (3) Does linguistic complexity influence synchrony perception? (4) And does synchrony perception explain individual variability in AV speech perception? These findings add value in deciding how AV synchronization should be balanced in hearing aid designs as well as in augmented / virtual reality devices.

AVbook, a high-frame-rate corpus of narrative audiovisual speech for investigating multimodal speech perception.

Seeing a speaker's face can help substantially with understanding their speech, particularly in challenging listening conditions. Research into the neurobiological mechanisms behind audiovisual integration has recently begun to employ continuous natural speech. However, these efforts are impeded by a lack of high-quality audiovisual recordings of a speaker narrating a longer text. Here, we seek to close this gap by developing AVbook, an audiovisual speech corpus designed for cognitive neuroscience studies and audiovisual speech recognition. The corpus consists of 3.6 h of audiovisual recordings of two speakers, one male and one female, each reading 59 passages from a narrative English text. The recordings were acquired at a high frame rate of 119.88 frames/s. The corpus includes phone-level alignment files and a set of multiple-choice questions to test attention to the different passages. We verified the efficacy of these questions in a pilot study. A short written summary is also provided for each recording. To enable audiovisual synchronization when presenting the stimuli, four videos of an electronic clapperboard were recorded with the corpus. The corpus is publicly available to support research into the neurobiology of audiovisual speech processing as well as the development of computer algorithms for audiovisual speech recognition.

Expression-tailored talking face generation with adaptive cross-modal weighting

The key of talking face generation is to synthesize the identity-preserving natural facial expressions with accurate audio-lip synchronization. To accomplish this, it requires to disentangle and fuse the latent features from multiple modalities, including the visual identity, facial expressions, and audio, etc. In this paper, we propose an end-to-end Expression-Tailored Generative Adversarial Network with Adaptive Cross-modal Weighting (ET-GAN-ACW). Different from previous talking face generation based on the identity image and audio, an expression video of arbitrary identity serves as the source in our system. On the one hand, multiple encoders are presented to disentangle the expression-tailored representation, audio-lip embedding, and face position localization in parallel. Additionally, instead of using a single image as the target identity, a multi-image identity encoder is proposed by exploring the different views of faces and merging them into a unified representation. These informative features from different modalities are then adaptively weighted and fused by the proposed Adaptive Cross-modal Weighting (ACW) mechanism. On the other hand, multiple discriminators are exploited to create the image-aware and video-aware realistic details, including a frame discriminator for the frame authenticity, and a spatial–temporal discriminator for the visual coherence of facial expression movements. Extensive quantitative evaluations on reconstruction error, identity preserving, expression retention, and audio-visual synchronization verify the superiority of our method. Qualitative results also demonstrate the effectiveness of our method in generating high-quality talking face videos.

A Timestamp-Independent Haptic–Visual Synchronization Method for Haptic-Based Interaction System

The booming haptic data significantly improve the users’ immersion during multimedia interaction. As a result, the study of a Haptic-based Interaction System has attracted the attention of the multimedia community. To construct such a system, a challenging task is the synchronization of multiple sensorial signals that is critical to the user experience. Despite audio-visual synchronization efforts, there is still a lack of a haptic-aware multimedia synchronization model. In this work, we propose a timestamp-independent synchronization for haptic–visual signal transmission. First, we exploit the sequential correlations during delivery and playback of a haptic–visual communication system. Second, we develop a key sample extraction of haptic signals based on the force feedback characteristics and a key frame extraction of visual signals based on deep-object detection. Third, we combine the key samples and frames to synchronize the corresponding haptic–visual signals. Without timestamps in the signal flow, the proposed method is still effective and more robust in complicated network conditions. Subjective evaluation also shows a significant improvement of user experience with the proposed method.

SyncTalkFace: Talking Face Generation with Precise Lip-Syncing via Audio-Lip Memory

The challenge of talking face generation from speech lies in aligning two different modal information, audio and video, such that the mouth region corresponds to input audio. Previous methods either exploit audio-visual representation learning or leverage intermediate structural information such as landmarks and 3D models. However, they struggle to synthesize fine details of the lips varying at the phoneme level as they do not sufficiently provide visual information of the lips at the video synthesis step. To overcome this limitation, our work proposes Audio-Lip Memory that brings in visual information of the mouth region corresponding to input audio and enforces fine-grained audio-visual coherence. It stores lip motion features from sequential ground truth images in the value memory and aligns them with corresponding audio features so that they can be retrieved using audio input at inference time. Therefore, using the retrieved lip motion features as visual hints, it can easily correlate audio with visual dynamics in the synthesis step. By analyzing the memory, we demonstrate that unique lip features are stored in each memory slot at the phoneme level, capturing subtle lip motion based on memory addressing. In addition, we introduce visual-visual synchronization loss which can enhance lip-syncing performance when used along with audio-visual synchronization loss in our model. Extensive experiments are performed to verify that our method generates high-quality video with mouth shapes that best align with the input audio, outperforming previous state-of-the-art methods.

Speech Driven Talking Face Generation From a Single Image and an Emotion Condition

Visual emotion expression plays an important role in audiovisual speech communication. In this work, we propose a novel approach to rendering visual emotion expression in speech-driven talking face generation. Specifically, we design an end-to-end talking face generation system that takes a speech utterance, a single face image, and a categorical emotion label as input to render a talking face video synchronized with the speech and expressing the conditioned emotion. Objective evaluation on image quality, audiovisual synchronization, and visual emotion expression shows that the proposed system outperforms a state-of-the-art baseline system. Subjective evaluation of visual emotion expression and video realness also demonstrates the superiority of the proposed system. Furthermore, we conduct a human emotion recognition pilot study using generated videos with mismatched emotions among the audio and visual modalities. Results show that humans respond to the visual modality more significantly than the audio modality on this task.

Soundscapes of war: the audio-visual performance of war by Shi'a militias in Iraq and Syria

Abstract This article sets out to bring sound and music to the field of visual studies in International Relations. It argues that IR largely has approached the visual field as if it was without sound; neglecting how audial landscapes frame and direct our interpretation of moving imagery. Sound and music contribute to making imagery intelligible to us, we ‘hear the pictures’ often without noticing. The audial can for instance articulate a visual absence, or blast visual signs, bring out certain emotional stages or subjects’ inner life. Audial frames steer us in distinct directions, they can mute the cries of the wounded in war, or amplify the sounds of joy of soldiers shooting in the air. To bring the audial and the visual analytically and empirically together, the article therefore proposes four key analytical themes: 1) the audial–visual frame, 2) point of view/point of audition, 3) modes of audio-visual synchronization and 4) aesthetics moods. These are applied to a study of ‘war music videos’ in Iraq and Syria made and circulated by Shi'a militias currently fighting there. Such war music videos, it is suggested, are not just artefacts of popular culture, but have become integral parts of how warfare is practiced today, and one that is shared by soldiers in the US and Europe. War music videos are performing war, just as they shape how war is known by spectators and participants alike.

Speech Enhancement for Multimodal Speaker Diarization System

Speaker diarization system identifies the speaker homogenous regions in those set of recordings where multiple speakers are present. It answers the question `who spoke when?'. The data set for speaker diarization usually consists of telephone, meetings, TV/ talk shows, broadcast news and other multi-speaker recordings. In this paper, we present the performance of our proposed multimodal speaker diarization system under noisy conditions. Two types of noises comprising additive white Gaussian noise (AWGN) and realistic environmental noise is used to evaluate the system. To mitigate the effect of noise, we propose to add an LSTM based speech enhancement block in our diarization pipeline. This block is trained on synthesized data set with more than 100 noise types to enhance the noisy speech. The enhanced speech is further used in multimodal speaker diarization system which utilizes a pre-trained audio-visual synchronization model to find the active speaker. High confidence active speaker segments are then used to train the speaker specific clusters on the enhanced speech. A subset of AMI corpus consisting of 5.4 h of recordings is used in this analysis. For AWGN, the LSTM model performance improvement is comparable with Wiener filter while in case of realistic environmental noise, the LSTM model improves significantly as compared to Wiener filter in terms of diarization error rate (DER).

Multimodal Speaker Diarization Using a Pre-Trained Audio-Visual Synchronization Model.

Speaker diarization systems aim to find ‘who spoke when?’ in multi-speaker recordings. The dataset usually consists of meetings, TV/talk shows, telephone and multi-party interaction recordings. In this paper, we propose a novel multimodal speaker diarization technique, which finds the active speaker through audio-visual synchronization model for diarization. A pre-trained audio-visual synchronization model is used to find the synchronization between a visible person and the respective audio. For that purpose, short video segments comprised of face-only regions are acquired using a face detection technique and are then fed to the pre-trained model. This model is a two streamed network which matches audio frames with their respective visual input segments. On the basis of high confidence video segments inferred by the model, the respective audio frames are used to train Gaussian mixture model (GMM)-based clusters. This method helps in generating speaker specific clusters with high probability. We tested our approach on a popular subset of AMI meeting corpus consisting of 5.4 h of recordings for audio and 5.8 h of different set of multimodal recordings. A significant improvement is noticed with the proposed method in term of DER when compared to conventional and fully supervised audio based speaker diarization. The results of the proposed technique are very close to the complex state-of-the art multimodal diarization which shows significance of such simple yet effective technique.

Realistic Speech-Driven Facial Animation with GANs

Speech-driven facial animation is the process that automatically synthesizes talking characters based on speech signals. The majority of work in this domain creates a mapping from audio features to visual features. This approach often requires post-processing using computer graphics techniques to produce realistic albeit subject dependent results. We present an end-to-end system that generates videos of a talking head, using only a still image of a person and an audio clip containing speech, without relying on handcrafted intermediate features. Our method generates videos which have (a) lip movements that are in sync with the audio and (b) natural facial expressions such as blinks and eyebrow movements. Our temporal GAN uses 3 discriminators focused on achieving detailed frames, audio-visual synchronization, and realistic expressions. We quantify the contribution of each component in our model using an ablation study and we provide insights into the latent representation of the model. The generated videos are evaluated based on sharpness, reconstruction quality, lip-reading accuracy, synchronization as well as their ability to generate natural blinks.

European Athletics Championship: Lessons from a Live, HDR, HFR, UHD, and Next-Generation Audio Sports Event

The European Broadcasting Union (EBU) and several of its member organizations undertook an ambitious trial over the summer period—a live, ultrahigh-definition, high dynamic range, high frame rate, wide color gamut, and next-generation audio trial at the European Athletics Championships in Berlin, Germany. The trial included four 2160p/100 cameras with live control of shading/racking, live vision mixing, live recording, live mixing of audio objects, audiovisual synchronization with audio objects, and live encoding. This paper looks at the lessons learned during the trial—what went well, what needed extra work, and what is needed to overcome current technical hurdles.

Audio–visual synchronization in reading while listening to texts: Effects on visual behavior and verbal learning

Reading while listening to texts (RWL) is a promising way to improve the learning benefits provided by a reading experience. In an exploratory study, we investigated the effect of synchronizing the highlighting of words (visual) with their auditory (speech) counterpart during a RWL task. Forty French children from 3rd to 5th grade read short stories in their native language while hearing the story spoken by a narrator. In the non-synchronized (S−) condition the text was written in black on a white background, whereas in the synchronized (S+) RWL, the text was written in grey and the words were dynamically written in black when they were aurally displayed, in a karaoke-like fashion. The children were then unexpectedly tested on their memory for the orthographic form and semantic category of pseudowords that were included in the stories. The effect of synchronizing was null in the orthographic task and negative in the semantic task. Children's preference was mainly for the S− condition, except for the poorest readers who tended to prefer the S+ condition. In addition, the children's eye movements were recorded during reading. Gaze was affected by synchronization, with fewer but longer fixations on words, and fewer regressive saccades in the S+ condition compared to the S− condition. Thus, the S+ condition presumably captured the children's attention toward the currently heard word, which forced the children to be strictly aligned with the oral modality.

Snow White and the Seven Dwarfs: Master Score by Walt Disney

Reviewed by: Snow White and the Seven Dwarfs: Master Score by Walt Disney Daniel Batchelder Walt Disney’s Snow White and the Seven Dwarfs: Master Score. Music by Frank E. Churchill, lyrics by Larry Morey, and additional underscore by Leigh Harline and Paul J. Smith. Limited edition digital replica. (Disney Music Legacy Libraries.) Milwaukee, WI: Hal Leonard Corporation, 2015. [Prologue, p. 4–5; score, p. 6–205; epilogue, p. 206–7; acknowledgements, p. 207. ISBN 978-14950-0872-6. $300.] Conventional musicological wisdom regards a score—say, that of a Beethoven symphony—as a blueprint of performable musical events. Scholars may subject these texts to stylistic analysis with an eye toward uncovering aspects of the composer’s musical thought process: why this chord, this rhythm, this instrument? Though certainly not without value, it is precisely this monolithic and quintessentially romantic approach that Ben Winters decries in a 2007 article on film-music scholarship (“Catching Dreams: Editing Film Scores for Publication,” Journal of the Royal Musical Association 132, no. 1 [2007]: 115–40). While the availability of published film scores has dramatically increased in recent years, Winters asserts that the auteurcentric methodology such documents enable distorts the emphasis on the composer’s creative autonomy by failing to account for the fact that scoring a film is a highly collaborative and fluid process. A film composer’s musical decisions are tempered by innumerable exigencies, from storyboard and sound-mixing concerns to editorial decisions that can leave feet of scored footage on the cutting room floor. This unavoidable indeterminacy, writes Winters (p. 120), “leaves us with a dilemma: how can we present a film score editorially in a way that both is useful for musicologists and yet acknowledges the often pluralistic, Text-like [in the Barthesian sense] complexities of the score?” An appealing solution to this dilemma has arrived in a surprising form. In late summer of 2015, the Walt Disney Company produced a beautifully bound, glossy facsimile of the manuscript “master score” to its 1937 animated film Snow White and the Seven Dwarfs. While it cannot claim to be the first published film score—or even the first publicly available animated film score, given Ries & Erler’s edition of Wolfgang Zeller’s Die Abenteuer des Prinzen Achmed from 1926 (Die Abenteuer des Prinzen Achmed: Musik zum Lotte Reiniger-Stummfilm, für kleines Orchester bearbeitet von Jens Schubbe, 5 vols. [Berlin: Ries & Erler, 2008])—this document is the first of its kind in many ways. Published by Hal Leonard and limited to a print run of 400 numbered copies, the edition was compiled and produced by a Disney-owned internal organization known as the Disney Music [End Page 157] Legacy Libraries (hereinafter, MLL). According to an explanatory two-page epilogue printed in the edition’s final pages, the MLL is an archival project dedicated to providing “a comprehensive and elegant resource solution for documenting and retrieving all phases of the [Disney studio’s] music scoring, contracting, and recording process” (p. 206). Since 2014, the MLL’s formidable parent company has exploited the project’s commercial potential with the Disney Music Emporium’s Legacy Collection, offering lavishly produced soundtrack recordings on compact disc and vinyl, complete with deleted songs and extensive liner notes (Walt Disney Records, “Disney Music Emporium: The Legacy Collection,” http://www.disneymusicemporium.com/the-legacy-collection [accessed 18 May 2016]). This score, however, signals the collection’s first foray into printed music. The publication’s use of the phrase “master score” compels some explanation. Though creating the score for a live-action film requires, as noted above, close and consistent collaboration between composers, editors, screenwriters, sound engineers, and the like, this process is magnified tenfold in the case of an animated film: every sound, utterance, and gesture must be carefully planned and organized in advance in order to ensure proper audiovisual synchronization. Though modern technology has streamlined and simplified this process, in the mid-1930s the creation of an eighty-three-minute animated film represented a gargantuan undertaking. By this time, however, the Disney studio had been at the forefront of the animation industry for nearly a decade, due in part to its constant experimenting with and implementation of new equipment...

3D Spatial Sound Systems Compatible with Human's Active Listening to Realize Rich High-Level <i>kansei</i> Information

In future communications with rich high-level kansei information, such as presence, verisimilitude, realism and naturalness, the role of sound is extremely important to enhance the quality and versatility of communications because sound itself can provide rich semantic and emotional information. Moreover, sound (auditory information) has good synergy effects with pictures (visual information). In this paper, we introduce our recent research results toward capturing and synthesizing comprehensive 3D sound space information as well as a high-definition 3D audio-visual display realizing strict audio-visual synchronization. We believe that these systems are useful to advance universal communications, which require particularly high-quality and versatile communications technologies for all.

Hearing voices then seeing lips: Fragmentation and renormalisation of subjective timing in the McGurk illusion

Due to physical and neural delays, the sight and sound of a person speaking causes a cachophony of asynchronous events in the brain. How can we still perceive them as simultaneous? Our converging evidence suggests that actually, we do not. Patient PH, with midbrain and auditory brainstem lesions, experiences voices leading lipmovements by approximately 200 ms. In temporal order judgements (TOJ) he experiences simultaneity only when voices physically lag lips. In contrast, he requires the opposite visual lag (again of about 200 ms) to experience the classic McGurk illusion (e.g., hearing ‘da’ when listening to /ba/ and watching lips say [ga]), consistent with pathological auditory slowing. These delays seem to be specific to speech stimuli. Is PH just an anomaly? Surprisingly, neuro-typical individual differences between temporal tuning of McGurk integration and TOJ are actually negatively correlated. Thus some people require a small auditory lead for optimal McG but an auditory lag for subjective simultaneity (like PH but not as extreme), while others show the opposite pattern. Evidently, any individual can concurrently experience the same external events as happening at different times. These dissociative patterns confirm that distinct mechanisms for audiovisual synchronization versus integration are each subject to different neural delays. To explain the apparent repulsion of their respective timings, we propose that multimodal synchronization is achieved by discounting the average neural event time within each modality. Lesions or individual differences which slow the propagation of neural signals will then attract the average, so that relatively undelayed neural signals will be experienced as occurring relatively early.

Other than Vérité: Sound and Moving Image in the Rock Music Documentaries of Peter Whitehead

When we think of music documentary and the emergence of rock in the 1960s, the classic fi lms of American cinema verite or direct cinema readily come to mind-What's Happening! The Beatles in the U.S.A. (Maysles brothers, US, 1964), Dont Look Back (D. A. Pennebaker, US, 1966), Monterey Pop (D. A. Pennebaker, US, 1968), Gimme Shelter (Maysles brothers, US, 1970), and Woodstock (Michael Wadleigh, US, 1970). These works have deservedly received great deal of critical attention. Unfortunately, their popularity suggests homogeneous approach to documentary during this era when, in fact, alternative practices fl ourished. Working coevally with Pennebaker, the Maysles brothers, and others, British fi lmmaker Peter Whitehead forged his own eclectic approach to documentary fi lms about pop u lar music. These groundbreaking works included documentaries on The Rolling Stones' 1965 tour of Ireland, The Beach Boys' 1966 visit to En gland, and early Pink Floyd in concert and in the studio. Because of legal obstacles that limited distribution, 1 Whitehead's fi lms are less well known than those of his American counterparts. However, viewers with sensibilities blunted by overexposure to the canon of direct cinema will discover in his fi lms refreshing take on the pop music scene of the de cade. Comparisons to fi lms such as Dont Look Back are expected,2 given the apparent similarities in subject and style. Yet, while sharing certain qualities with direct cinema, Whitehead's work remains profoundly dissimilar. This difference derives in part from the geo graph i cal distance between the United States and Britain, which allowed Whitehead to develop working practices unhampered by the obsessive polemics of the Americans. Because he drew inspiration from sources both selective and wide- ranging-from the Eu ro pe an art fi lm auteurs such as Bergman and Godard and from his training as newsreel cameraman- his diverse body of work refuses facile pigeonholing. Ultimately, Whitehead is bricoleur of cinema, determined to get the job done with the tools hand in the time allowed. I focus here on Whitehead's approach to audiovisual synchronization, for I believe this aspect clearly distinguishes his work from that of his American contemporaries and also illustrates his improvisatory manner of working. Four of Whitehead's music documentaries are addressed, each of which has its own mode of treating the relationship between moving images, speech, and music. I demonstrate that such diverse approaches preclude any adherence to the doctrine of verite. The Doctrine Although the practitioners of direct cinema did not conform to the caricature typically painted of them, it is nonetheless accurate to claim that they pursued naturalism as repre sen ta tional practice- especially as it pertained to audiovisual components.3 For Leacock, Pennebaker, Drew, Maysles, and company, portable sync sound represented the Holy Grail.4 To reject it apostasy. Yet sync sound sought after without critically examining its value. Why it so desirable? Perhaps perceptible disjunction between what we see and what we hear makes us wonder whether some tampering might have occurred in that gap. In other words, hiatus between sound and image presents an opportunity for the fi lmmaker to intervene. The power to capture reality unawares has persisted as an ideal, but as one seldom attained. We may imagine pure form of direct cinema, but the practice of fi lmmaking inevitably necessitates compromise.5 As Joe McElhaney observes, the Maysles brothers violated the codes of direct cinema at almost every turn.6 Nevertheless, these fi lmmakers continued to pledge themselves to a documentary cinema in which truth and reality are simply out there in the world, waiting to be captured on fi lm.7 Whitehead admits for time, he too was swept up little by that, but he soon realized that the notion of the objective documentary totally wrong and a total lie. …