Preface These Skill Level Descriptions are primarily intended to serve as guidelines for use in government settings.
Audio translation is the process of rendering live or recorded speech in the source language to a written text in the target language. It is a cross between interpretation (speech-to-speech) and translation (written text-to-written text), and requires a skill set that includes not only language but also the ability to overcome input interference.
Audio translation directly from live speech is not likely to result in a full translation.  There are also a variety of tasks that may not require a full translation of the audio source, such as summarizing, identifying significant items, listening for details, or preparing analytical reports for a specific user or purpose.
This document covers only products that transfer audio materials as fully as possible into another language. They are prepared from recorded speech, and are often used for legal purposes by many government agencies, particularly courts, which require either certified translations based on transcripts, or transcript and translation side by side.
A successful audio translation is herein defined as one that fully and accurately conveys the content and meaning of the source language in a script format, and reflects the style, register, and cultural content of the source message, without additions and omissions. From the standpoint of the user, the translation must also meet the prescribed specifications and be completed in a timely manner.
Competence in two languages is necessary but not sufficient for any audio translation task. The audio translator must be able not only to (1) comprehend the source language; and (2) write intelligibly and idiomatically in the target language; but also (3) choose the equivalent expression in the target language that both fully conveys and best matches the meaning intended in the source (congruity judgment).

Most importantly, the audio translator must also be able to overcome factors impeding comprehension of the source. These may be related to input quality, speech character or discourse content.
Factors that obliterate, diminish or interfere with input are often termed “unfavorable conditions,” and may include low volume, background noise, gaps, multiple speakers, overlapping exchanges, interruptions, and fast or faulty delivery.
Factors related to speech character impede comprehension and include language that is accented, substandard, slang, very colloquial, or substantially deviating from the norm.
Factors related to content hinder understanding of the source message and may include complex subject matter, coded or deceptive language, technical jargon, place and name references, and out-of-context statements.
Resolving these issues requires highly developed linguistic and non-linguistic skills, particularly since audio translators are passive listeners, without the opportunity to request clarification.
As with any language transfer, knowledge of socio-cultural factors and familiarity with the subject matter are necessary. Audio translators must also continuously update their knowledge of the language in order to handle the latest colloquial variations. Analytical and research skills allow the individual to proceed methodically, using language tools and on-line aids. Techniques related to the workplace must be acquired, including the use of special equipment and software.
In view of the many requirements and pitfalls involved in audio translation, extreme care must be exercised in hiring audio translators or assigning tasks to them. To do otherwise entails the risk that imprecise or even wrong information will be conveyed.
Therefore, assessing the ability to perform audio translation presents a special challenge. Tests that measure listening comprehension in the source language and writing skills in the target language should be considered screening tools only. Additional job-related performance testing that reflects real life tasks is absolutely necessary.
Individuals should be assigned to tasks within their performance levels. To facilitate this correspondence, the Skill Level Descriptions that follow are divided into three bands: Minimal Performance (levels 0+ to 1+), Limited Performance (levels 2 and 2+), and Professional Performance (levels 3 to 5).

Back to Top

Minimal Performance Levels 0+ to 1+ are characterized by weaknesses in all of the requisite audio translation skills.

Limited Performance Levels 2 to 2+ are characterized by weaknesses in one or more of the requisite audio translation skills.

It is at Professional Level 3 that all necessary skills begin to align and enable production of reasonably accurate translations of audio materials.

At Professional Level 4 (and above) competence and expertise combine to produce accurate and reliable translations of a variety of audio materials.
Examples of tasks and texts appropriate for each level are provided, and each level implies control of all functions at the lower levels. The “plus level” designation is used to describe performance which substantially exceeds the next lower skill level but for any reason does not fully meet the criteria for the next higher level.

For further specification, reference must be made to the ILR Skill Level Descriptions for Translation Performance and Interpretation Performance.



Back to Top
Level 5 (Master Professional Performance): Able to produce fully successful translations of audio materials consistently and reliably. Can overcome, to the extent possible, virtually all factors impeding comprehension of the source.

Level 4+ (Advanced Professional Performance Plus): Able to produce successful translations of audio materials consistently. Can overcome, to the extent possible, nearly all factors impeding comprehension of the source, and can be relied upon to complete assignments well within deadlines.

Level 4 (Advanced Professional Performance): Able to produce full and accurate translations of audio materials, generally reflecting style, register, and cultural context in most respects. Can overcome, to the extent possible, most unfavorable conditions and other factors impeding comprehension of the source. Except for passages that may be particularly unclear, repeated listening of the recording is often not necessary, enabling the individual to proceed at a fast pace and normally meet deadlines.

Level 3+ (Professional Performance Plus): Able to produce accurate translations of most audio materials. Can normally render jargon, slang, accented, colloquial, regional, and substandard speech. Can usually overcome most unfavorable conditions and other factors impeding comprehension of the source.

Level 3 (Professional Performance): Able to produce reasonably accurate translations of conversations that exhibit some complexity and deal with topics outside everyday matters. Can usually render jargon, slang, and speech that is colloquial, substandard, or regional. Able to capture most nuances, idioms, and cultural allusions, reflecting the source register appropriately. Can often, but not always, overcome many unfavorable conditions and other factors impeding comprehension of the source. Completes assignments in a timely manner.

Level 2+ (Limited Working Performance Plus):
Able to render straightforward conversations mostly accurately. Demonstrates emerging ability to transfer slang, colloquial, substandard, regional, or coded language, and can overcome some unfavorable conditions, such as multiple speakers and overlapping exchanges.

Level 2 (Limited Working Performance): Able to render with some accuracy straightforward everyday conversations on concrete matters, and topic-specific information if familiar with the subject matter. Can sometimes overcome sporadic unfavorable conditions, such as background noise, after listening repeatedly to the recording.

Level 1+ (Minimal Performance Plus): Able to render, but not always accurately, simple routine conversations from mostly clear recordings. Can rarely overcome any factors that impede comprehension even after listening repeatedly to the recording.

Level 1 (Minimal Performance): Able to transfer short and very simple routine conversations, delivered in the variety of the language with which the individual is familiar. Recordings must be clear and without any factors impeding comprehension. Accuracy is haphazard.

Level 0+(Memorized Performance):
Able to transfer isolated words and/or phrases from very clear recordings.

Level 0 (No Performance): No functional ability to transfer information from one language to another.


Approved by the Interagency Language Roundtable on 28 September 2012.

Back to Top