The Importance of the Often Neglected Forensic Transcription/Translation

In my time working as a court interpreter here in Alabama, there’s one mistake I see attorneys and courts make time and again that can lead (and have led) to miscarriages of justice: failing to get a forensic transcription/translation done of foreign language audio recordings. If you’re not a court interpreter, you may be asking yourself what on earth I’m talking about. Let me give you an example. Let’s say that an attorney is appointed to represent a Limited English Proficient defendant in a criminal trial. During discovery, she receives a copy of the recording of her client’s interrogation, which led to his arrest. The interrogation was done either using a bilingual speaker (usually a friend, family member, or bilingual police staff) as an interpreter or by a bilingual law enforcement officer, which means 50 to 100% of the recording is in a foreign language that the attorney does not understand. Other recordings regularly seen in cases with foreign-language speakers are wiretaps of conversations between a suspect and confidential informant or undercover officer, cellphone footage of police encounters, or conversations between co-conspirators.

When such evidence is in play, as the court interpreter, I see one of two things happen: If there’s someone acting as an interpreter in the recording, it is assumed that that person interpreted competently and no one bothers to have the recording assessed by an expert. If a party does want to call the interpretation in the audio into question or if the recording is strictly in a foreign language, the court will sometimes request that the court interpreters render a live interpretation of the audio being played in the courtroom. Most interpreters, myself included, will refuse to do so for reasons we’ll look at in this post. Other interpreters will try to do a live rendition that, nine times out of ten, won’t meet the high threshold of our oath to render a faithful, complete, and accurate interpretation. At no point does anyone treat this recording as an important piece of forensic evidence or ask if there may be a professional who specializes in working with it.

My hope today is to inform officers of the court that, indeed, there are experts who can provide the exact services needed in these cases. I want court personnel, and court interpreters who may be pressured into doing a live interpretation of a recording, to understand what a transcription/translation is, why it is so important, and the risks of not hiring the right professional to do the job.

What is a transcription/translation?

Transcription/Translation (TT) is a highly specialized discipline that is a hybrid of court interpreting and legal translation. TT differs from transcripts you may see in other audiovisual industries, in that the forensic TT expert must capture all sound elements in the recording: stammering, repetitions, hedges, grammatical errors, incorrect word usage, sounds, background noise, different speakers, swear words, slang, overlapping speakers, etc. Like court interpreters, they cannot clean up what they hear and they must try to capture as much of what is heard on tape as possible. The expert can’t decide what sounds on the tape merit being documented. That dog barking in the background, the speaker stammering, parties whispering, a muted microphone, or the music playing could all be of forensic importance to the case, so all of it must be reflected in the final transcript. The Transcription Translation Expert (TTE) must then take that finalized transcription and provide a written translation into English that best preserves the meaning of the original, including niche (think code words used in the drug or human trafficking trade) or regional/cultural references and meanings. A professional transcription/translation will be formatted so that the parties that are relying on it can follow along with the recording. The end product must be something that the TTE can defend on the witness stand as a qualified expert. It is excruciating, long, and tedious work, even with the help of the best transcription and dictation software available on the market.

How can transcription/translations help in a case involving a foreign language speaker?

As the National Association of Judiciary Interpreters & Translators rightly points out in their position paper on transcription/translation, transcripts are considered an “Aid to Understanding”, or alternatively, Opinion Evidence. I couldn’t put it better than they did, so I will quote from their position paper directly: “In either instance, a transcript is produced by the prosecution to aide in processing sound that may not be readily accessible to the unassisted ear of jurors, parties, and the Court…When any portion of the sound file is challenged, in the absence of a transcription/translation, the only source of reference is the official record, which will have to be read every time reference is made to a specific portion of the recording. If someone challenges the use of a specific word, there is no transcription to reference for the word in question. Aside from a reading from the official record, the only other options are to rely on the memory of the interpreter, who may or may not be present in the courtroom during all proceedings, or to listen to the sound file each time a word or phrase is challenged. This method is not expeditious. For this reason, a transcription translation that stands up to scrutiny in the courtroom and meets legal evidentiary standards should always be used.”[1]

Why can’t the court interpreters just do a live interpretation of the recording?

Due to a lack of knowledge about TT and in an effort to cut costs, many officers of the court will erroneously assume that they can deal with foreign language recordings by simply using court interpreters to render live interpretations of the audio, usually in the consecutive mode of interpreting. While I understand how someone could arrive at this conclusion, it is wrong for two reasons. First, while some court interpreters may be trained in TT or have knowledge of what is needed to do produce a good TT, they will not have all of these tools at their disposal to successfully deliver an English rendition of the foreign language recording. In order to successfully produce a good transcription/translation, a professional needs knowledge of the transcription process, time, technology, and research tools, none of which are present when doing a live rendition. As NAJIT, yet again, says beautifully, “A forensic psychiatrist would be remiss, indeed disqualified, if he provided an expert opinion on a patient based on seeing the patient for the first time in the court room. Along the same lines, forensic TT requires expertise, time, and technology to perform the work required to an adequate standard.”[2]

Second, because they do not have these tools at our disposal and, therefore, can only produce a mediocre rendition of the audio at best, court interpreters put themselves in danger of violating their oath to render a complete, full, and faithfully accurate interpretation for the record. We serve at the pleasure of the court and have a duty to protect the record. We simply cannot do that when we’re asked to do an on-site live interpretation of recordings, no matter how crystal clear the audio is. Some may try and argue that we can do on-site interpretations of recordings, comparing it to Sight Translation. The difference here is that sight translation is within our parameters as court interpreters, where interpreting evidentiary recordings is not. We are trained and tested on our sight translation skills, but not on interpreting sound files. The sound file used in the court certification exam is also not comparable, because those are meant to mimic everyday court procedures interpreters encounter on the job. They DO NOT include sound files mimicking audiovisual evidence. Additionally, we have specific limits placed on sight translations. We are limited to only sight translating short documents. We have to decline anything lengthier or more technical and recommend that the court get a professional translation done instead. I would argue that onsite interpretations of recordings are more comparable to sight translations of lengthier documents in that they should not be done and need to be referred to other experts.

I would also argue that by interpreting evidence presented at a hearing or trial, court interpreters are putting their neutrality at stake. This is because, especially in instances where they are asked to interpret audio where someone else served as an interpreter (for example, in a recording of an interrogation), interpreters may be asked to opine on the quality of the interpretation heard in the recording. At that point, we step out of our role as the court interpreter and into the role of an expert witness for the party seeking to admit or dismiss that piece of evidence. When this happens, our neutrality is compromised for the rest of the proceeding. Even in civil litigation, where interpreters are specifically hired by one of the parties, acting as an expert in this capacity could call into question your neutrality. This conflict is easily resolved by bringing in an outside forensic TT expert, leaving the assigned court interpreters’ neutrality intact and free to interpret court proceedings without further complications.

What does the transcription/translation process look like?

You may be asking yourself what this work entails, especially to justify the high cost of these services. Let’s say you have a piece of audio recording that you want transcribed and translated and you’ve located a potential expert. Great! First, the expert must get a sense for what the assignment entails and decide if there’s any conflict of interest. Next, they’ll need a copy of the recording. You, the client, must keep the original to preserve the chain of custody. The copy should be of the highest quality available, which can be difficult since most recording devices used in court cases aren’t top of the line. While it need not be perfect, make sure the copy is as good as possible to lessen the risk of having portions deemed “unintelligible”.

Next, you’ll need to look at the purpose of the transcription/translation. Are you looking for a rough draft/summary of the sound file or do you need a complete TT? It’s fine to request a summary for limited informational purposes, but keep in mind that this won’t be a certified TT for use in court and that the expert will make a note identifying the final product as a draft or summary. Also, keep in mind that “in the case of a summary being requested, the [expert] must advise the client that producing a summary calls for the [expert] to make a judgement call as to what constitutes important information and what does not. This lies outside the [expert’s] area of expertise.”[3] Caveat emptor: you’re requesting a summary at the risk of missing details that would only be available in a complete certified transcription/translation.

Once the client’s request is clear, the main TTE will review the recording to assess how much time he or she will need to complete the project, if the deadline is realistic, or if one or more additional TTE teams will need to be brought in to tackle the project. While the rule of thumb is one audio minute equals one hour of work, this can vary depending on the quality of the recording and complexity of the speech recorded. After assessing the recording, the TTE will outline for the client what time and resources they will need to complete the assignment. The TTE will also reach out to their editor and other teammates (if needed) to make sure they are available for the assignment.

If the client gives the green light, the main TTE will begin transcribing the sound file. Forensic transcription is very tedious, even with technology that helps cut down on work time. Because of the nature of forensic transcription/translations, we cannot just stick an audio file in a transcribing program and rely on the final product. In fact, many encrypted audio file formats typically seen in a court case are not compatible with this type of software, meaning more manual work for the TTE. Even when software can be used, the TTE will still need to go back through the audio and make sure that the transcription is correct and complete. This also means additional research to verify that they are hearing what the speakers are saying correctly, including code words, slang, geographical markers (streets, cities, landmarks, etc.), errors, and misspeaks. When the TTE finishes their first draft of the transcription, they’ll send that draft to the editor for revisions. The editor will send back the draft with their notes, which the TTE will look over. If he or she disagrees with any of the editor’s suggestions, the TTE and editor will work together to resolve the dispute.

When the transcription is finalized, the translation process begins. Before beginning the translation, the TTE will look over the transcription and research any terminology that could prove problematic. For certain niche topics, they may need to consult an outside expert. For example, if in a recording there is a high amount of Nicaraguan gang slang, I would need to consult a colleague familiar with that subsect of terminology. After completing their glossary, the TTE will start the first draft of the translation. Once the first draft is done, the TTE will send it to their editor. The editing process can be just as extensive as translating. The editor will need to make sure that the translation corresponds correctly to that portion of the transcription, that nothing has been omitted, and that the translation accurately reflects the text of the original. The editor will provide edits and notes, and the TTE and editor will send it back and forth, until they have a finalized version of the translation, which the TTE will certify.

The TTE will then turn in the final product to the client: a side-by-side transcription/translation of the audio recording, with an attached glossary and index (if needed). At the top of the Forensic TT, they will identify the recording and also provide a legend so that the reader can easily follow along as they go through it. This, ladies and gentleman, is why one minute of audio can require an hour or more of work.

What should I look for in a prospective forensic TT expert?

Once again, NAJIT does a wonderful job outlining[4] what types of skills and credentials you should look at when hiring a prospective expert. They rightly point out that “not all interpreters are adept at transforming the spoken word into written text with the accuracy required in the legal setting. By the same token, professional translators may lack the training to accurately transform live recorded extemporaneous speech into written form.” Ideally, you’re looking for a professional who is trained and familiar with the nuances of court interpreting, as well as written translation technique. As a former attorney, I would say that practitioners should make sure that they hire someone who can pass the voir dire procedure and be qualified as an expert on the witness stand. Remember: whomever you hire may have to defend their transcription/translation under oath.

The first step in narrowing down potential candidates is looking at their credentials. The National Center for State Courts, the credentialing body of most state court interpreters, requires a credentialed transcription/translation expert for legal transcription/translation work. While there is no specific certification for TT, you’ll need to get an extensive history of the expert’s training and specify who granted them their credentials. There are three existing credentials that you could look for in your potential expert. Translators are certified by the American Translators Association. Keep in mind that the ATA exam tests general translation skills in various English/Foreign Language pairings. Their exams are not bidirectional, meaning that they test specifically from one source language into one target language. For example, if a translator wants to be certified to translate to and from English & Spanish, she will need to sit two separate exams: one testing her skills from Spanish into English and a second testing from English into Spanish.

For court interpreters, there are two types of certification tests. The Administrative Office of US Courts provides a certification exam for Spanish court interpreters only.[5] State courts certify their interpreters either through the National Center for State Courts or through their own internal examination (ex. New York State). The state level certification examination is available in Spanish and Languages Other Than Spanish (LOTS). Unlike the ATA exam, court interpreting exams are bidirectional, as they test court interpreting skills both into and out of English and the foreign language. A potential expert who holds one of these three credentials is a good starting point, but you’ll want to consider their training and work experience extensively. For languages that do not have a certification exam, the process will be even more complicated and you should consult with an expert who provides forensic transcription/translation services on how to approach this situation.

In your search, you must also keep in mind what skills prospective experts must have in addition to credentials. According to NAJIT,[6] some of the skills these experts should have include:

  • Acute hearing
  • Native-quality knowledge of languages
  • Understanding of cultural factors
  • Expertise in recognition of language registers
  • Formal higher education
  • Analytical skills
  • Attention to detail
  • Knowledge of research methodology
  • Ethical expertise
  • Problem-solving skills
  • Neutrality
  • Awareness of forensic testimony requirements
  • Ability to self-monitor and correct
  • Openness to third-party review
  • Knowledge of technical tools
  • Openness to new technology and methods

From this list, I want to highlight the skills mentioned relating to acting as an expert witness in a case. Just like members of other professions, not all interpreters & translators are adept at being expert witnesses. It is important that you hire someone that knows what to expect as a potential expert witness. This should not discourage you from hiring someone newer to TT, since we all have to start somewhere, but it should drive home the point that you can’t just pluck out any court interpreter to do this work. Given all of the credentials, skills, and experience that are required of forensic TT experts, combined with how exhaustive this work is, it should come as no surprise that these services are not cheap. In fact, inexpensive transcription/translation services are a red flag. Remember: this is work usually done in teams of two or more, where one audio minute of recording is equivalent to one hour of work. While some of these experts may be open to negotiating fees for special circumstances or indigent clients, be wary of something that sounds too good to be true. As many civil litigation attorneys will tell you, you get the expert you pay for.

Who benefits from a professional transcription/translation?

Right now, there is not a lot of official policy on the books on using professional qualified interpreters in the investigation phase of a case, be it in criminal or civil court. While there are great position papers providing guidance on this,[7]professional interpreters typically aren’t brought in until the preliminary hearing phase. To put it shortly, you never know what you’re going to get when it comes to language services used to communicate with LEP suspects, witnesses, and victims during an investigation. Once you get to the hearings phase of a case, the window to look at this evidence begins to close and those court interpreters assigned to interpret in proceedings must not be used to interpret and assess what is heard in the recordings. This puts the accusing and defending parties at risk of being both perpetrators and victims of miscarriages of justice.

A well-done forensic TT of audiovisual recordings can help all parties rest assure that they have the necessary facts to proceed with the case. Prosecutors and defense attorneys will have all of the evidence needed to present their best case and uphold their professional duties. Judges will also be able to make more informed rulings. LEPs going through the justice system will also be one step closer to equal access to the courts. It is a win-win for all parties involved. I say this because I don’t want to give the impression that transcription/translations can only be used to dismiss charges or appeal decisions. A lack of forensic TT can also harm the State’s chances of proving a case or a victim’s chances of successfully seeking justice. Remember: when the proper language professionals aren’t used everyone is harmed. The defendant is harmed. Victims are harmed. Families and friends, the community, the courts, taxpayers, and even the State are harmed. Everyone loses. Getting a forensic transcription/translation of audiovisual evidence is going to take more time and require more of a financial investment up front, but in the long run, it can prevent so much unnecessary harm.

Do I provide Forensic Transcription/Translation services?

The short answer is no, not right now.

While there is no hard and fast rule on when a court interpreter is ready to provide these services, it’s important to read the position papers provided by our professional organization and reflect on what you need to be ready to act as an expert. You may arrive at a different conclusion regarding the career milestones you need to meet before working as a TT expert. That’s completely fine. I’m just offering my reflections to help my colleagues.

Even though I am certified in two states as a Spanish court interpreter, I made the call a while back not to provide these services until I was certain that I was in a place professionally to deliver a solid product and defend my work under scrutiny as an expert witness. I have no qualms about testifying as a witness, but I want to have all the goods to back me up under a tough interrogation. First, because I work with Spanish, I want to hold the highest certification standard for Spanish court interpreters: federal certification. Additionally, I’d also like the ATA’s Spanish > English certification to further back up the translation portion of my work product. Second, I needed to take a continuing education course to learn the fundamentals of forensic transcription/translation. I took the one offered by Castillo Language Services, which I highly recommend. Finally, I need to have a reliable network of colleagues that have met the same credentialing standards and are willing to work with me as a part of a forensic transcription/translation team as my editors, collaborators on projects that require multiple transcription/translation teams, and as resources for the different regional variations of Spanish and niche terminology that can pop up in court cases. My professional goal is always to take care of the trust that is being instilled in me by the client, courts, and LEPs. I will only start offering this service once I meet these benchmarks that will allow me to fulfill this goal.

[1] National Association of Judiciary Interpreters and Translators Position Paper: “Onsite Simultaneous Interpretation of a Sound File is Not Recommended”.

[2] See Footnote 1.

[3] National Association of Judiciary Interpreters and Translators Position Paper: “General Guidelines and Requirements for Transcription Translation in a Legal Setting for Users and Practitioners”.

[4]See Footnote 3.

[5] Note: The Administrative Office of US Courts formerly offered the court certification exam in Haitian Creole and Navajo, but those exams have since been retired.

[6]See Footnote 1.

[7] National Association of Judiciary Interpreters and Translators Position Paper: “Language Assistance for Law Enforcement”.