Using NICTs for Media Interpreting Training: Bringing Interpreter-Mediated TV News and Radio Interviews to the Classroom

This paper discusses how the impact of the revolution of NICTs on interpreters' work in the media can be transposed to the interpreting classroom. This article focuses on the use of sound and video editing software by interpreters for producing prerecorded interpreter-mediated news features for TV and live dialogue radio interpreting. The main purpose is two-fold: 1) to acknowledge the potential of NICTs for interpreter training in a practical context; and 2) to build a pedagogical model based on the pioneering practice of media outlets such as ARTE, the French-German broadcasting company, and RTVE, the Spanish public broadcaster.


Introduction
Media interpreting (MI), especially on TV, has been the focus of an increasing body of research over the last 20 years, with an exponential increase in research literature in the last decade. This is most likely due to the fact that scholars have gained easier access to interpreter-mediated data, as well as to the high degree of exposure that MI has provided to practicing interpreters (cf. Castillo 2015b, Jiménez Serrano 2011, Pöchhacker 2007. However, it can be argued that the growing body of scholarly work on MI has not, as yet, had a significant impact on the training of future interpreters, based on the scant literature on the use of interpreter-mediated broadcasts for training purposes. Although there have been attempts to create media-based corpora to take advantage of interpreter-mediated broadcasts for training purposes (Castillo 2015a, Jiménez Serrano 2011, Straniero Sergio 2011, Sandrelli 2012, no project has yet come to light that has systematised such data and produced a pedagogical model to make the most of this type of interpreting for training purposes, along the lines of the model developed by de Manuel Jerez (2003. Neither has MI been included in interpreter training curricula at undergraduate or postgraduate levels, as suggested by Castillo (2015a).
The purpose of this paper is to discuss the impact of the revolution of NICTs (New Information and Communication Technologies) on interpreters' work in the media, most notably on TV and radio. Furthermore, to show how interpreter-mediated TV and radio broadcasts can be used for interpreting training purposes with the enhancement of new technologies, i.e., audio editing tools, within a pedagogical framework which takes into account the organisational, interactional and discursive aspects of dialogue interpreting in radio settings (2015a).
Media interpreting (MI) is a growing field within interpreting research (e.g. Castillo 2015b, Falbo 2011) and the profession (e.g. AIIC 2012). Yet, in spite of the growing number of interpreter-mediated broadcasts available on TV, radio and online, their potential for training purposes is still underdeveloped, or at least not documented in the literature. This article specifically explores TV news reports including short interviews, and interpreter-mediated live radio interviews in which dialogue interpreting is provided. In my previous research (Castillo 2015a), I concluded that radio interviews can bridge a common gap in training dialogue interpreting: the challenge of providing interpreting students with original dialogues, as well as creating authentic settings for this modality in which to practice their interpreting skills. This claim can also be applied to TV interpreting; TV and radio still constitute the main core of the so-called mass media.
Building on previous work in interpreter training action research (de Manuel Jerez 2003, Sandrelli andde Manuel Jerez 2007), the project presented here offers a practical application in consecutive, simultaneous and dialogue interpreting classes using edited interpreter-mediated broadcasts from the French-German broadcasting company ARTE and RNE, the Spanish public radio service. Editing for training purposes consists of the following: in the case of TV news, video/audio software (Pro Tools or similar free software) is used to produce voiceovers of news pieces which include interviews or statements in a foreign language, following ARTE's now customary practice (Krone 2010) and technical specifications (ARTE G.E.I.E. 2012). In the case of radio interviews, audio software (Audacity) is used to remove interpreters' utterances from the broadcast piece and leaving those spaces blank for students to interpret as if it were live interpreting, following RNE's Radio 3 practice. Based on observational ethnographic research and the broadcast information provided by the media institution, the interpreting conditions are replicated in the classroom by showing images of the setting where the interviews take place, re-enacting the situational arrangements and the on/off-air levels of interaction.
Special emphasis is put on class exercises which tackle key organisational aspects of different live radio interview settings, e.g., studio interaction, studio-to-studio interaction and studio-to-remote location interaction.
Examples of actual exercises at different stages of interpreter training (undergraduate, MSc, MI workshops for practitioners) which I have conducted in the last 10 years are provided, paving the way for a project to systematise such data and produce a pedagogical model to make the most of MI for training purposes. Technologies as diverse as traditional media (radio), web-based media platforms (podcasting) and audio software interact to enhance the students' learning experience. Braun (2006:1) argues that: the ongoing spread of information and communication technologies along with growing multilingualism and efforts of social inclusion (access to the media for all) has led to changes in communication practices, which have also had repercussions on the practice of interpreting at the beginning of the 21st century.

NICTs and interpreter training
Technological developments such as teleconferencing technologies, webcasting and podcasting have resulted in "some relatively new forms of interpreting and have created additional and/or novel tasks for interpreters" (Braun, 2006:2). There is no doubt that NICTs have revolutionised the ways interpreting is taught. As de Manuel Jerez argues (2003), the amount of material that can be used for both class and self-study interpreting practice has grown exponentially since we entered the era of the internet and digital media. Amongst the main advantages and potential of new technologies for interpreter training, he mentions the following: This access to real interpreter-mediated material has been made even easier since the so-called YouTube (and similar platforms) revolution as well as TV and radio on demand, which offers the possibility to access vast amounts of data that was otherwise inaccessible just 15 years ago. In spite of these enormous advantages, the introduction of more authentic and easily accessible data in interpreter training programmes is not without its challenges. As Bao (2015) argues: As it is often assumed that students themselves are responsible for their own practice, especially graduate students who are supposed to know how to study, practice is very often left in the hands of the students. But the truth is that it may take a while for students, especially in their first semester, to find the right approach to practice. Without specific assignments given by the instructor, students often do not know what to practice with and how. For the training to produce the desired results, it is useful to integrate student self-study and practice into the curriculum and provide guidance accordingly. In some programs, students are given credit for self-study, and well-structured practice materials are provided. (Bao, 2015:413-414) This warning call is shared by , who substantiate their argument with education theories and subsequently suggest a greater implication of trainers in integrating NICT-based interpreter training platforms which have been developed in the last 20 years and are being used with some degree of success in interpreting training programmes, mostly for conference interpreting (Berber 2008 (Braun et al., 2013:95) and focus "rather exclusively on self-study and individual practice" (2013:97). Nevertheless, these platforms tend to be limited in terms of the number of languages and language combinations, although with successful examples of progress being made in recent years, such as the Speech Repository. They also lack an integration of computer-assisted interpreter training (CAIT) packages "which offer digital content along with bespoke learner and authoring functions for an interpreting context" (Braun et al., 2013:97). Although this represents significant progress in interpreter training and these packages are increasingly being used in the classroom (Berber 2008), the pedagogical model behind them has fostered self-guided and autonomous learning. This only satisfies cognitive constructivist principles of learning and omits "principles of a social constructivist approach to learning (Vygotsky 1978)" which highlight "the importance of the social embedding of learning for the process of knowledge construction" (Braun et al., 2013:97). Following this approach would involve, according to Kiraly, a combination of "authentic situated action, the collaborative construction of knowledge, and personal experience" ( regarding platforms/projects such as the Speech Repository and ORCIT by acknowledging that the selection of AV material for interpreting training purposes entails certain risks and recommending that the following three aspects be taken into consideration when selecting AV material for interpreting training purposes, to ensure coherence: Professional reality.
How technological advances can help us bring this reality into interpreting laboratories.
How and when to introduce material thus obtained into interpreter training.
Some innovative projects using NICTs have also been carried out in other fields of interpreting practice, although few results have been reported in research as to how effective these projects are. However, it is worth noting that they do exist: in business interpreting Hansen and Shlisinger (2007)  In the field of public serviceor communityinterpreting, Tymczyńska (2009)  Adam Mickiewicz University (AMU), Poznań, that is: "pedagogy must drive the technology and not vice versa. Technology is an important tool, but as with any tool, it is useful only if one knows how to use it advantageously" (Gómez & Weinreb 2002: 643). One of the main advantages of CMS, according to Tymczyńska, is that: Auditory Learners who learn through listening to audio and video materials; Tactile/Kinaesthetic Learners who profit from doing interactive online exercises and self-assessment tests. (Tymczyńska, 2009:156) Although most of these projects are ongoing, 2 what seems clear is that the turn of the 21st century has seen a move towards a social constructivist and communicative approach (cf. Sandrelli 2015), as evidenced by Kiraly (2000), de Manuel Jerez (2003, Sandrelli and de Manuel Jerez (2007), (Hansen and Shlisinger 2007) and Tymczyńska (2009). NICTs give interpreter trainers an unprecedented opportunity, moving beyond traditional teaching strategies such as staging dialogues in class or playing pre-recorded speeches to be interpreted simultaneously or consecutively by students. Using NICTs, trainers can gather examples of interpretations that have been broadcast on TV, radio or online and edit them in a way that allows students to step into the shoes of the original interpreter and create their own interpretation, under the same or similar conditions. This does not only mean that students can compare their own work with that of professional media interpreters, but also increases their motivation by bringing authentic examples of interpreting into the classroom.

NICTs and MI training
When it comes to MI, however, research and literature on the pedagogy of MI is even scarcer and there have been few attempts to bring increasing knowledge on MI into the classroom in a systematised way (cf. Castillo 2015b). Yet, Berber's (2008) hypothesis that more and more interpreter trainers use interpreter-mediated broadcast material in their classes (most often in conference interpreting classes), as colleagues from the UK and Spain have confirmed to me informally, seems to be not far from reality. The success rate of these practices, however, has not been studied empirically. On the other hand, some skills and challenges can be common to interpreting across modalities and fields of practice (Gile 2009), and although their weight may vary from one modality/field of practice to the next, training exercises can be adapted to the specificity of these fields of practice on the basis of thorough institutional, discursive and communication research. In this sense, role plays and interpreting drills have been -and still are -common practice in interpreter training (particularly in dialogue interpreting in the field of public service interpreting, and conference interpreting) and they are being used systematically with a certain degree of success (e.g. Wadensjö 2014, Zhong 2010). Yet they still run the risk of not being perceived by trainees as sufficiently authentic, 4 which, in turn may diminish both trainee motivation to carry out the exercises and trainers' credibility when it comes to using these types of exercises adapted to MI.
This is not to say that MI can only be taught through exercises that use actual broadcast material, but it is a warning that any replication of real-life situations should be backed by a thorough knowledge of the media, including production practices and what happens on and off the broadcast. Furthermore, replicating broadcasting studio conditions in the interpreting lab is subject to potential technical limitations in the latter. However, with field and ethnographic research, it is possible to get to know MI production conditions, as well as how MI is used to serve language transfer purposes in media broadcasts.  Gieve and Norton (2007) offer one of the first attempts to classify the strategies for dealing with linguistic difference on broadcast television systematically (including interpreter-mediated and non-interpreter-mediated encounters). For the purposes of this article, a key contribution of Gieve and Norton's study is a model which critically describes "how the production as a whole articulates and mediates the relationship between the protagonist, audience and the FL speaker as broadcast, through pre-, during-and post-filming strategies." In the second stage of their descriptive analysis, Gieve and Norton (2007:194-206) identify strategies for the representation of linguistic difference that they summarise in the following  (Gieve and Norton, 2007:207).
Beyond the political and ideological implications of these strategies (Gieve and Norton, 2007:208), a key contribution of this study for interpreter training is that it breaks away from a monolithic view of TV interpreting as a practice that typically takes place live, in either consecutive or simultaneous mode, therefore opening up opportunities for a wider range of practices that through research, could be turned into training exercises and skills.
Castillo and Comte (2010 and 2011) take a step further in the exploration of interpreting practices in TV and radio production with the research-based documentaries 5 The table excludes news reports and political interviews (Gieve and Norton, 2007:206).
Although the categorization does not draw on exhaustive quantitative analysis, the authors mark the most commonly found strategies in bold.  (2010), which offer audio-visual evidence of the creation and production mechanisms in a multilingual television programme requiring interpreters and in live radio interviews, respectively. Comte (2010 and show the collaborative nature of media production, which must be taken into consideration when it comes to interpreter-mediated events. Moreover, as Castillo (2012:85) argues following his data analysis: Media stakeholders' views and approaches to interpreting sometimes collide or need to be negotiated, and it is the level of language and interpreting awareness, experience and collaborativeness of the production team that ultimately shape the way interpreting is performed and broadcast.
As opposed to Gieve and Norton (2007:208), who suggest an 'ideal approach' which recognises that linguistic differences do exist, in predictable as well as unpredictable contexts, Castillo and Comte (2011) and Castillo (2012)  to categorise interpreter-mediated TV events and interviews (Alexieva 1999 and 2001, respectively) did not look into these aspects systematically and it could be argued that the map of MI practices was incomplete, with radio interpreting not featuring at all.  With such models to account for MI practices, more realistic training exercises based on interpreter-mediated broadcasts are possible, particularly thanks to the use of NICTs that give way to innovative training methods and exercises, as argued in the previous section. The following sections focus on how this can be put into practice.

Building on interpreter-mediated broadcasts: ARTE and RTVE
As we have seen above, the body of research on MI is growing, largely focusing on descriptive accounts of what MI entails and how it works in different broadcasters and/or countries, using both spoken and sign languages (e.g. Mizuno 1997, Molina Vallecillo 2002, Darwish 2006, Shibahara 2009, Tsuruta 2011, Jiménez Serrano 2011, Castillo 2015b. Studies have investigated the challenges that interpreters face and specific skills that are required (e.g. Daly 1985, Kurz 1997, Kurz and Bros-Brann 1996, Straniero Sergio 2003, as well as looking into organisational aspects and interactional and discursive aspects (e.g. Wadensjö 2000, 2008, Straniero Sergio 1999, Katan and Straniero Sergio 2001, Gieve and Norton 2007, Castillo 2015a).
These studies have one methodological aspect in common: they all look at real interpreter-mediated practice, whether it is at the level of whole TV/radio stations, specific programmes, events or interactions in the media. There have also even been attempts to build corpora that can be used to find patterns, as a basis for potential training, or both (see specifically the volume edited by Falbo and Straniero Sergio 2011, but also Falbo 2012, Sandrelli 2012. As Falbo (2011: XVI) argues: "studying TI through corpora is a further step towards new interrogation paradigms, new methodological challenges and, hopefully, more ecologically comparable results." Therefore, corpus-based approaches to MI are not only possible, but desirable if training in MI aims to be representative of real practices.
For the purpose of this study, I draw on two main public broadcasters (ARTE and RTVE) with a long history of interpreting practice in their broadcasts and whose organisational and interactive aspects have been researched systematically in order to 118 lend authenticity to the training exercises suggested below. The samples of interpreting available from these broadcasters are limited to four languages (English, French, German and Spanish), but it is hoped that they can be replicated and extended to other languages according to trainers' needs.

Methodology
Based on the social constructive approaches reviewed earlier, which advocate situated learning issuing from action-research methods, this paper offers a case study of a small sample of interpreted news and interviews on TV and radio, to be interpreted and recorded by students using free and accessible software, such as Audacity or Pro Tools. This model encourages students' in-class as well as guided independent practice, and it requires a prolonged interaction with NICTs. It also focuses on key MI skills such as use of the voice and engaging with common news and interview discourses, which all form part of a set of training activities which motivate students to practice interpreting and improve their delivery skills from the early stages of their training.
This model also attempts to address the aims and programme guidelines established by the EMCI (European Masters in Conference Interpreting) agreement (2014), specifically when it comes to emphasizing the need for systematic training and guiding individual and group self-directed study (EMCI 2017) 5. Suggested practices of TV and Radio interpreting using NICTs: stages and dynamics The exercises are grouped into two main blocks by interpreting modality. The following description highlights the key features that make each type of clips suitable for interpreting practice, thus permitting trainers to identify the relevant criteria when researching their own exercise material. As general preparation for both types of exercises, students are provided with background information on the broadcasting institutions, the TV channel and radio station, and the programmes where the interpretedmediated interviews take place, with special emphasis on analysing the respective communicative ethos (Scannell 1988), as a way of anticipating discursive and interactional aspects that may be relevant to the interpretation. Face-to-face, multi-party interview. For providing feedback to students a two-fold system is proposed: first, a qualitative feedback session, using the interview transcription, either with the students' recordings or via a post-exercise discussion; second, a more reflective feedback session which consists of playing the original audio broadcast with the interpreter's rendition and allowing students to comment and compare the broadcast interpreting with their own to integrate their performance in a situated practice and, thus, encouraging them to see the value and implications of exercises based on authentic practice.

Concluding remarks
With the suggested exercises, presented as case studies, several broadcast events with a diversity of situational arrangements and interactional challenges have been presented in a systematised way.
As the two proposed exercises and the methodology for creating them show, the preparation of such materials requires a great deal of time and effort from the trainer.
In fact, the work presented here is the result of nearly ten years of research on TV and radio broadcasting under different projects and over 14 years of teaching interpreting.
This professional experience has convinced me that in order to pave the way for future MI training and interpreting training overall, two steps are of fundamental importance: first is working with interdisciplinary groups (including interpreters, scholars, media stakeholders and professionals,); and second, establishing action-research projects which have the time and resources to produce pedagogically sound training materials grounded in professional practice. aim to meet the demand for highly-qualified interpreters. It is not enough for a professional media interpreter to have proficient interpreting skills; they also need to have a thorough understanding of the media and the new technologies involved in producing interpreter-mediated broadcasts. It is this gap that this article aims to fill.