IRB Protocol Language for Recording + Transcription (Ready-to-Use Wording)

If your study uses audio or video recording, your IRB will usually want clear protocol language on how you record, transcribe, protect, and destroy data. The easiest way to write it is to break the process into steps: recording, transcription, de-identification, storage, access, and retention. Below, you’ll find ready-to-use wording researchers can adapt for the Methods or Procedures section, plus a checklist of common IRB attachments.

• Use clear protocol language for each stage of recording and transcription.
• Match your consent form, data security plan, and third-party vendors to the protocol.
• State who can access identifiable files, where they are stored, and when they will be destroyed.
• Remove or replace direct identifiers in transcripts as early as possible.
• Keep your wording specific enough for review, but flexible enough for real study workflows.

What should IRB protocol language for recording and transcription include?

IRB protocol language for recording and transcription should explain what you will record, why recording is needed, who will have access, how transcription will happen, and how you will protect participant privacy. It should also describe de-identification, storage location, security controls, retention period, and destruction methods.

Your protocol should align with your consent materials and any institutional rules on sensitive or identifiable data. If your research involves protected health information, review the HIPAA Privacy Rule and your institution’s privacy guidance.

Ready-to-use IRB protocol language for Methods or Procedures

You can adapt the sample text below to fit interviews, focus groups, usability sessions, oral histories, or other recorded activities. Replace bracketed text with your study details and remove anything that does not apply.

1. Recording procedures

Sample wording:

"With participant permission, the research team will audio-record [and/or video-record] study sessions to support accurate capture of participant responses. Recording will occur using [device/platform], and only the portion of the interaction necessary for study purposes will be recorded."

"At the start of each session, the researcher will remind participants that recording is optional if applicable under the approved consent process, explain the purpose of the recording, and confirm whether the participant agrees to proceed. If a participant declines recording, the researcher will [take written notes / use an approved alternative procedure]."

"Recordings will be labeled with a study ID or code rather than the participant’s name whenever feasible. Any file that contains direct identifiers will be handled as identifiable research data."

2. Transcription method

Sample wording:

"Recordings will be transcribed to create a text record for analysis. Transcription will be completed by [members of the research team / an approved transcription provider / a combination of both]."

"If a third-party transcription provider is used, only the minimum necessary data will be shared for transcription, and the provider will be used in accordance with institutional requirements, contractual terms, and approved study procedures. The research team will review completed transcripts for accuracy and remove or replace direct identifiers as described below."

"If the study uses automated tools to assist with transcription, the research team will review and edit the output before analysis. Automated output will not be treated as final without human review."

If you are comparing manual and automated workflows, it may help to define the difference between automated transcription and human-reviewed transcription in your protocol and internal SOPs.

3. De-identification of transcripts

Sample wording:

"Following transcription, the research team will de-identify transcripts by removing or replacing direct identifiers, such as names, contact details, exact addresses, employer names, school names, medical record numbers, or other information that could directly identify a participant. The team may also generalize certain indirect identifiers when needed to reduce re-identification risk."

"A master list linking participant names to study IDs will be stored separately from transcripts and recordings in a restricted-access location. Only authorized study personnel will have access to the linkage file."

"When quotations are used in reports or publications, the research team will review them to reduce the risk that a participant could be identified through context."

4. Storage and security

Sample wording:

"Digital recordings, transcripts, and linkage files will be stored on [institution-approved encrypted storage / secure server / approved cloud environment] protected by password-controlled access. Study data will not be stored on personal devices unless those devices are encrypted and approved under institutional policy."

"Data transferred between approved systems or approved service providers will use institutionally approved secure methods. The research team will follow the institution’s data security requirements for identifiable research data."

"If paper notes are created during or after recording, they will be stored in a locked cabinet or other restricted-access physical location when not in use."

Many institutions expect security language to reflect established controls such as encryption, access limits, and secure disposal. For general guidance, researchers can review the NIST media sanitization guidance when describing deletion or destruction methods.

5. Access controls

Sample wording:

"Access to identifiable recordings, transcripts, and linkage files will be limited to authorized members of the research team who require access to perform study duties. Team members will be trained on study procedures and confidentiality requirements before accessing study data."

"If external vendors or transcription personnel are involved, their access will be limited to the data needed to complete assigned tasks, and access will end when those tasks are complete or when the agreement terminates."

"The principal investigator will oversee access permissions and will review data handling practices throughout the study."

6. Retention and destruction plan

Sample wording:

"Identifiable recordings will be retained only for as long as needed to complete transcription, quality review, and any approved verification steps, unless a longer retention period is required by the sponsor, institution, or law. After that point, recordings will be securely deleted or destroyed using institutionally approved methods."

"De-identified transcripts and coded study data will be retained for [insert retention period] in accordance with sponsor, institutional, and legal requirements. The linkage file will be destroyed when it is no longer needed for study purposes, unless retention is required by an approved protocol, sponsor, or law."

"If a participant withdraws, the research team will handle existing recordings and transcripts in accordance with the consent form, institutional policy, and the extent to which data have already been incorporated into analysis."

How to tailor the wording to your study

Do not paste sample language into your protocol unchanged. IRBs look for wording that matches the real workflow, the consent form, and the tools your team actually plans to use.

• State whether you will record audio, video, screen activity, or a mix.
• Name the platform or type of device if your IRB expects that level of detail.
• Explain whether recording is optional or required for participation.
• Describe who creates transcripts and whether a vendor is involved.
• Say when de-identification happens and who does it.
• List where files are stored and what security controls apply.
• Define when recordings are deleted and whether transcripts remain coded or de-identified.

If your study includes multilingual interviews, you may also need to describe whether translation occurs before or after de-identification and whether a separate audio translation service or text translation workflow will be used.

Common mistakes that slow IRB review

Many delays happen because the protocol is too vague. Small gaps can create major review questions when the IRB tries to understand where identifiable data will go.

• Saying recordings will be stored “securely” without naming the approved system or control type.
• Forgetting to mention who can access raw recordings.
• Not explaining whether third-party transcription providers will receive identifiable data.
• Using different retention periods across the protocol, consent form, and data security plan.
• Promising full anonymity when recordings or linkage files make the data identifiable or coded.
• Leaving out what happens if a participant refuses recording.
• Failing to describe how direct identifiers will be removed from transcripts.

Before submission, compare every mention of recording and transcription across your protocol packet. The wording does not need to be identical, but it should be consistent.

IRB attachment checklist for recording and transcription studies

Most studies that record and transcribe data need more than protocol language alone. A short checklist can help you catch missing documents before submission.

Common attachments to prepare

• Updated consent form or script that explains recording and transcription.
• Recruitment materials, if they mention recording or remote platforms.
• Data security plan or data management plan.
• Third-party agreement, vendor review, or institutional approval for any outside transcription provider.
• Confidentiality language for research staff or transcription personnel, if required by your institution.
• Interview guide, focus group guide, or session script.
• Template for study IDs or coding structure, if your IRB requests it.
• Retention and destruction schedule, if submitted separately.
• HIPAA authorization or waiver materials, if applicable.
• Materials for translated consent or multilingual transcription workflows, if applicable.

Quick consistency check before submission

• Does the consent form clearly say that recording will occur?
• Does the protocol explain why recording is necessary?
• Do the protocol and consent form match on retention and destruction timing?
• Have you disclosed any vendor, platform, or external data transfer that will occur?
• Have you explained whether transcripts will be de-identified?
• Have you described what happens if a participant declines recording or later withdraws?

Choosing a transcription approach for your protocol

Your protocol should describe a transcription method that fits your risk level, timeline, and data type. The right choice depends on how sensitive the material is, how much cleanup you can do in-house, and what your institution allows.

• Research team transcription: useful when you want direct control over identifiable data and coding.
• Third-party human transcription: useful when you need a separate workflow and can support it with approved agreements and controls.
• Automated-first workflow with human review: useful when your protocol allows it and your team can validate and correct the output.

When writing the protocol, avoid broad claims that one method is always best. Instead, explain your chosen method, how you will review the transcript, and how you will protect participant data during each step.

If you need a practical starting point for evaluating options, review service details, workflows, and transcription pricing before finalizing your protocol language and budget.

Common questions

Do I need to say who will transcribe the recordings?

Yes, in most cases you should identify whether transcripts will be created by the research team, a vendor, or both. If a third party is involved, say how data sharing will be limited and governed.

Should I promise anonymity in a study with recordings?

No, not unless your IRB agrees that the data are truly anonymous. If you collect recordings or keep a linkage file, the safer description is usually confidential, coded, or de-identified after collection.

When should de-identification happen?

Do it as early as your workflow allows. Your protocol should say whether identifiers are removed during transcription, during transcript review, or immediately after transcript verification.

Can I keep recordings after transcription is complete?

Possibly, but your protocol should explain why. If you do not need them for quality checks, audit needs, training approved by the IRB, or another stated purpose, many teams choose a shorter retention period for raw recordings.

What if a participant refuses to be recorded?

Your protocol should state the alternative, such as researcher notes, unless recording is required for the study procedure. Your consent process should make this point clear.

Do I need separate language for video versus audio?

Often yes, because video can increase identifiability. If you capture faces, screens, locations, or documents on camera, your protocol should say so and explain the added protections.

Final tip: make the protocol easy to review

The best IRB protocol language for recording and transcription is simple, specific, and consistent across the full submission packet. If reviewers can quickly see what you will record, who will handle it, how you will protect it, and when you will delete it, the review process is usually much smoother.

If your team needs help turning recordings into usable text while fitting a documented workflow, GoTranscript provides the right solutions, including professional transcription services.