Language Learning & Technology
Vol. 4, No. 2, September 2000, pp. 31-36

REVIEW OF Roberto's Restaurant CD-ROM



Virtual ConversationsTM Language Programs: Roberto's Restaurant CD-ROM 
Publisher:  Interactive Drama Inc.
7900 Wisconsin Ave., Suite 200
Bethesda, MD 20814
Phone: (800) 489-4547, (301) 654-0676
Fax: (301) 657-9174
ISBN: 1-892601-13-3
Platform:  PC only, Windows 95/98 or NT4
Will run on Macintosh with Virtual PC*NOTE*
Minimum hardware requirements:  Pentium processor 100 with 16MB of RAM, 15MB of free hard drive space, 4x CD-ROM drive, SVGA display (800x600, 256 color), Sound Blaster compatible sound card, computer speakers, mouse. A microphone is included with purchase. 
Support offered:  (800) 489-4547 and
Target language:  Spanish 
Target audience:  Not stated, but it appears to be designed for advanced intermediate or advanced adult students 
Price:  $89.95 

Reviewed by Robert Blake, University of California, Davis



Roberto's Restaurant CD-ROM is a stand-alone PC software package which purports to offer the user a simulated face-to-face conversation (Restaurant Mode) and a parallel practice session (Instructor Mode) where responses can be matched against an accent evaluation routine. At its heart, this CD employs a modified "imitate and memorize" (MIM-MEM) approach, common to the pre-communicative teaching methodologies, which is designed to familiarize the user with the rhythm and sound of Spanish. There is no grammar instruction or formal vocabulary presentation. The user's exposure to "real-life" situations is limited to the restaurant domain. Roberto's Restaurant appears designed to address the problem of language recovery and maintenance rather than to provide beginning language instruction, and would thus best be used as a curricular supplement for high intermediate and advanced levels. While the accent evaluation function is impressive, the program's lack of sound language pedagogy mitigates its overall effectiveness as an instructional aid.


Roberto's Restaurant automatically installs the appropriate files when opened for the first time. The user should pay particular attention to the sound card directions that come with the program in order to insure that the voice recording feature will work properly. Once the CD-ROM is loaded and the button labeled "Roberto's Restaurant" has been clicked, the user sees a short video clip with lively Latin American music of people entering a restaurant (unfortunately, this QTmovie cannot be stopped; it has to run its full course every time the program is accessed).

The user is then confronted by two, side-by-side, video head shots (see Figure 1). On the left, an unnamed young woman, the instructor, waits patiently; on the right, Roberto, the owner of a Mexican restaurant, and the user's interlocutor in the simulated conversation, smiles and occasionally looks furtively about the room.

Figure 1. The Instructor and Roberto

- 32 -

The user chooses either the Instructor Mode or the Restaurant Mode by clicking on the corresponding image. In the Instructor Mode (see Figure 2), the user can practice all of the conversational responses needed to "interact" with Roberto in the Restaurant Mode. This would seem the logical way to begin the program, but nothing prevents the user from plunging directly into the simulated conversation with Roberto.

Figure 2. The Instructor Mode

In the Instructor Mode, the user can choose to work on a variety of restaurant topics including greetings, ordering drinks, main courses, desserts, coffee, helping another couple order, and leaving. Suggested responses to Roberto's prompts are listed in the "Phrases" field with translations readily available in the adjacent field on the right, if desired. When any given response is selected with the mouse, the instructor will immediately model it in the video window with a standard Latin American accent. The user can then repeat and record that response for playback and comparison with the native speaker's recording.

The CD-ROM's most distinctive feature is the evaluation of the learner's accent in Spanish. By clicking on the "Recognize" button, the user's pronunciation is rated (from poor to excellent) according to how well it matches a composite energy-wave pattern. Although it would undoubtedly be possible to "fool" this speech-recognition algorithm with different regional accents or individual voice differences, this reviewer found the results satisfactory and even impressive at times, despite being limited to the small inventory of stipulated responses. From this screen, the user can then move on to the simulated dialogue with Roberto (see Figure 3).

- 33 -

Figure 3. The Restaurant Mode

In the Restaurant Mode, the video character, Roberto, first addresses the user about the last topic practiced in the Instructor Mode. Nonetheless, the button on the bottom left allows the user to choose other topics, just as is the case in the Instructor Mode. After Roberto speaks, three prospective responses successively appear in a field at the bottom middle of the screen, at about two-second intervals. The user clicks on one of these prompts to "answer" Roberto and can read the prompt out loud, but there is no recording option in the Restaurant Mode. Once one of the three prompts is clicked, Roberto will respond appropriately.*NOTE* The company has described this feature as "intelligent user prompts" (Harless, Zier, & Duncan, 1999, p. 314). However, it was impossible for this reviewer to evaluate how "intelligent" this prompting system really is because the prompts, although different for each of Roberto's utterances, are always the same on each use of the program, and Roberto's retort to each response is also predictably the same.

Other options given at the bottom right of the screen allow the user to hear Roberto's utterance again, display it in writing under the video window, or review the entire dialogue for any given topic. The user can return to the Instructor Mode or exit the program at any time.


This program offers a unique accent evaluation algorithm and a promising prompting system within a limited language pedagogical design. If language recovery and maintenance is truly the company's goal, they should state this clearly in their literature. Calling Roberto's Restaurant a simulated dialogue is accurate, but referring to as a virtual face-to-face dialogue in dramatic real-life situations is less accurate, given that the learner's participation is limited to clicking on one of three prefabricated written responses. In the Instructor Mode, the user is similarly restricted to repeating canned responses, rather than engaging in a meaningful conversation. Such repetition, while it may still have its place in the curriculum, is not a reflection of current SLA practices (see Gass 1997, especially chapter 7). The user will find little in the way of meaningful interaction, negotiation of meaning, or task-based learning in this CD, which is based more on traditional imitation, translation, and memorization. Resolving grammatical and lexical confusions is entirely dependent on reading sentence translations, as no reference materials or explanations are provided.

- 34 -

From a Spanish-language pedagogical perspective, there are several unusual aspects to the dialogues themselves. Roberto frequently alternates between formal and informal terms of address (usted and tú, respectively) within a single topic while talking to the same user "¿Qué deseas comer?" (informal) and "¿Por qué no le ofrezco. . .?" (formal). There is no sound pedagogical reason for this; in fact, it could be considered a communicative deficiency in most Spanish-speaking societies. Roberto also exhibits occasional non-standard use of the subjunctive (e.g., "Esperamos que le gusta. . ."), a fact that the company describes in their literature as a sign of authentic, natural, and conversational Spanish: Presenting learners with nonstandard input doesn't strike this reviewer as following any SLA goal of providing either positive or negative input. It's difficult to accept the rationale for including ill-formed sentences/

In more than a few user responses, English is used. In the Instructor Mode, the video character doesn't model these English phrases, but rather just says, "English." It seems strange that the user would want to use English halfway into the conversation (around dessert time) with Roberto. His response to "Do you have flan?" is predictably "No le entiendo" (I don't understand you.), but this hardly contributes to the learning experience in any meaningful way.

Also, about halfway into the virtual dialogue with Roberto, a monolingual English-speaking couple on their honeymoon comes into the restaurant. The user ends up having to order for them, lending an added complication to the plot. Strangely, the wife says she wants chicken (which the restaurant surely has) but the husband and user order a hamburger for her. From a cross-cultural perspective, all the stereotypes seem to come out. The couple represent the ugly Americans: they want paella (a dish typical of Spain) and American hamburgers, but this is supposedly a Mexican restaurant set in Mexico. Although the waiter explains this, the user demands to see the owner, Roberto, which could be interpreted as a typical sign of American arrogance. Stereotypes of Mexicans are also well represented: the interaction with the American couple is followed by a dirty-fork incident and a fly-in-the-salad adventure. We also "learn" that Mexicans have good teeth because they eat lots of beans. While there certainly exists an improvised feel to the dialogue, a little more cultural sensitivity and pedagogical scripting and would not be amiss.


The program's accent evaluation feature and the intelligent prompting could provide useful tools for computer-based language learning, especially if they were integrated into a clearly articulated language pedagogy. Roberto's Restaurant is a welcome supplement to the Spanish curriculum, but it needs to be used in a carefully defined learning context where the grammar and vocabulary have previously been introduced. Adding more lexical domains other than that of the restaurant would also greatly enhance its usefulness. Still, it was fun to match and rate responses against the native speaker models and some of Roberto's retorts will be very entertaining, if not motivating, for the more advanced language learner.

- 35 -


Robert Blake (Ph.D., University of Texas at Austin) is professor of Spanish and Director of the Second Language Acquisition Institute at the University of California, Davis. He was the academic consultant for Nuevos Destinos, the principal author of Al corriente: Curso Intermedio de español, 3rd edition, and is presently part of a team of authors about to publish Tesoros CD-ROM: Curso Introductorio. Currently, he is researching the effects of integrating CMC into the foreign-language curriculum.



Gass, S. (1997). Input, Interaction, and the Second Language Learner, Mahwah, NJ: Lawrence Erlbaum Associates.

Harless, W. G., Zier, M. A., & Duncan, R. C. (1999). Virtual Dialogues with Native Speakers: The Evaluation of an Interactive Multimedia Method. CALICO Journal 16(3), 313-337.

- 36-

Home | About LLT | Subscribe | Information for Contributors | Masthead | Archives