Expert report review
How do you challenge the other side's forensic audio expert report?
A review of the opposing audio report starts with the file: which recording the examiner received, whether it was the original, and what was done to it. The reviewer then checks the tests against the SWGDE best practices for digital audio authentication and the voice comparison against the standards a federal court applied in Angleton. Absent a stipulation or order, a rebuttal report is due within 30 days after the other side's disclosure.

What is a review of the other side's forensic audio report?
A written analysis by another audio examiner of the file, the tests and the conclusions in the opposing report. Counsel can use the reviewer as a consultant or disclose the reviewer as a rebuttal expert.
An audio report in litigation answers one or more of three questions: whether a recording was edited, whose voice is on it, or what was said. The review takes the opposing report apart along the same lines. It asks which file the examiner received, what was done to it, which tests were run, what each test found and whether the conclusion follows from the results.
Rule 26 of the Federal Rules of Civil Procedure gives the reviewer one of two roles. A testifying expert is disclosed under Rule 26(a)(2)(A) and, if retained, signs a written report under Rule 26(a)(2)(B). An expert retained only to prepare for trial and not expected to testify falls under Rule 26(b)(4)(D): ordinarily the other side may not discover the facts known or opinions held by that expert, except under Rule 35(b) or on a showing of exceptional circumstances under which it is impracticable to obtain the same facts or opinions by other means.
A consulting reviewer reads the report, lists the gaps, drafts document requests and prepares questions for the deposition and for cross-examination. A testifying reviewer does the same work and then signs a rebuttal report that contains all opinions with their basis and reasons, the facts or data considered, exhibits, qualifications with publications from the previous 10 years, testimony in the previous four years and the compensation.
The two roles can be combined in sequence. Counsel can start with a consulting review and decide, before the rebuttal deadline, whether the reviewer will testify.
When is a rebuttal audio report due, and what if it is late?
Absent a stipulation or court order, within 30 days after the other side's disclosure. A party that misses a disclosure deadline may lose the right to use that witness, unless the failure was substantially justified or harmless.
Rule 26(a)(2)(D) says a party must make expert disclosures at the times and in the sequence the court orders. Absent a stipulation or order, the main disclosures are due at least 90 days before the date set for trial or for the case to be ready for trial. Evidence "intended solely to contradict or rebut evidence on the same subject matter identified by another party" is due within 30 days after the other party's disclosure.
For audio work the 30 days are tight. The reviewer needs the opposing report with its exhibits, the file the examiner used in the form the examiner used it, and often the recording device or exemplar recordings. Each of these may require a document request or a meet-and-confer before anything is analyzed. A request sent on day 25 leaves no time for the analysis.
Rule 37(c)(1) sets the cost of a missed disclosure: if a party fails to provide information or identify a witness as Rule 26(a) or (e) requires, the party is not allowed to use that information or witness to supply evidence on a motion, at a hearing or at a trial, unless the failure was substantially justified or is harmless.
The opposing examiner may also change the report. Under Rule 26(e)(2) the duty to supplement extends both to information included in the report and to information given during the expert's deposition, and any additions or changes must be disclosed by the time the pretrial disclosures under Rule 26(a)(3) are due. The reviewer checks each supplement against the original report.
What should counsel request for the reviewer?
The report with its exhibits, the questioned file in its native form with metadata, the hash values recorded by the examiner, the recording device or its model, any exemplar or known recordings, and the examiner's notes on each test.
Rule 26(a)(2)(B)(ii) makes the report list "the facts or data considered by the witness in forming" the opinions, and item (iii) requires "any exhibits that will be used to summarize or support them." For an audio report the facts or data are files and measurements: the questioned recording, any copies the examiner made, exemplar recordings, known voice samples, spectrograms, waveform displays and the results of each test.
Rule 34(a)(1)(A) reaches "sound recordings" among the documents and electronically stored information a party may request. Rule 34(b)(1)(C) lets the requesting party "specify the form or forms in which electronically stored information is to be produced." A request for the native file with its metadata, rather than a converted copy, is made at this point. If the request does not specify a form, Rule 34(b)(2)(E)(ii) requires production in the form in which the information is ordinarily maintained or in a reasonably usable form.
The request list follows the steps in SWGDE Best Practices for Digital Audio Authentication (document 15-a-001, version 1.3, September 20, 2018): the item submitted and whether it is the original recording; the hash of that item and of the working copy; the assessment data (container, encoding, channels, bit rate and sample rate, duration, encoded date and time); the test plan and the scope agreed with the submitter; exemplar recordings made on the same make and model of recorder; and the documentation of procedures, data, results and decision criteria.
For a voice comparison the list adds the known recordings of each person compared, their source and date, the segments used, the features measured and the software with its settings.
Did the examiner work on the original file?
The reviewer checks what the examiner received and how it was preserved. SWGDE treats a transcoded or re-encoded version as not an original and has the examiner hash the original and work on a copy.
Section 3.1 of the SWGDE guide has the examiner clarify the request and confirm whether the submitted item is the original recording, and it states that transcoded or re-encoded versions are not originals. A file sent through a messaging app, exported from a phone into another format, or trimmed by the party before production fits that description.
Section 2 defines handling of the original: protect it and hash it if possible. Appendix A describes two bit-stream images, one archived and one used as the working copy. The reviewer looks for the hash values in the report and compares them with the file produced in discovery. A report that records no hash leaves no way to prove which file was examined.
For voice comparison the same point appears in older standards. Bruce E. Koenig of the FBI published in 1993 a table of minimum conditions used by the International Association for Identification and the FBI: both required the original recording (Crime Laboratory Digest, vol. 20, no. 4, Table 2).
If the examiner worked from a copy, the review states what the copy can and cannot support. SWGDE's analyses of file format and structure (4.4.1-4.4.2) read the container and encoding of the item examined, and on a re-encoded copy those belong to the copy.
What does the reviewer check in an authentication report?
Whether the examiner ran the analyses SWGDE lists, documented each one, compared the file with exemplars from the same device model, and relied on no single test.
Section 4.1 of the SWGDE guide requires that "the test plan, procedures, data, and results shall be documented," together with the decision criteria, and adds that no single analysis shall be relied upon solely: all applicable tests are run so that results can be checked against each other. A report that rests on one spectrogram or on listening alone departs from that rule.
The guide sets an order of work. The examiner first assesses the file (4.2) and performs critical listening "without applying any processing" (4.3). Global analyses follow (4.4): file format and structure, quantization level, DC offset, long-term spectral analysis, compression level, MDCT analysis and electrical network frequency analysis (4.4.8). Local analyses (4.5) look at specific points: waveform, signal power, spectrogram and butt-splice detection.
Device verification and exemplars come next (4.6 and 4.7). The examiner compares the evidence with recordings made on the same make and model of recorder. If the opposing examiner had no exemplar recordings, the reviewer asks how the examiner knew which header values, sample rates and file structures the device produces.
The reviewer reads the test results against the conclusion. If one test finds a discontinuity and another finds a file structure that matches the device's exemplars, the report has to explain how the examiner weighed them.
What conclusions may an authentication report state?
One of three outcomes in section 5.2 of the SWGDE guide: consistent with an original, inconclusive, or inconsistent with an original. The guide tells examiners to avoid claims of absolute certainty.
Section 5.2 lists the outcomes: "Consistent with an original," "Inconclusive" and "Inconsistent with an original." Discontinuities and alterations found during the examination are included in the report with an explanation of their cause. A report that says "no edits" without stating which tests were run and what each found does not meet that standard.
Section 5.2.1 reads: "No scientific inquiry, including those in forensics, produce a result of absolute certainty." Absolute language is to be avoided unless the examiner is describing known alterations or deletions. The reviewer marks every absolute statement in the opposing report.
Section 5.3 points examiners to ASTM Practice E620 for reporting opinions and lists required items, among them a "Description of the evidence and its chain of custody" and the "Data and reasoning relied upon to come to conclusion(s)." A conclusion with no data and reasoning behind it is the gap that General Electric Co. v. Joiner, decided December 15, 1997, called "simply too great an analytical gap between the data and the opinion proffered."
The 2023 committee note to Rule 702 makes a similar point for all experts: "each expert opinion must stay within the bounds of what can be concluded from a reliable application of the expert's basis and methodology," and forensic experts should avoid assertions of absolute or one hundred percent certainty when the method is subjective.
What does the reviewer check in a voice comparison report?
The known recordings and their conditions, the amount of usable speech, the method and its accuracy, how the examiner arrived at the conclusion, and what the examiner was told about the suspect.
United States v. Angleton, 269 F. Supp. 2d 892, decided June 26, 2003, gives a list. Judge Rosenthal of the Southern District of Texas excluded a defense examiner's aural spectrographic comparison. The standards of the International Association for Identification and the American Board of Recorded Evidence called for a list of usable words on both recordings, similar and dissimilar; the examiner listed only dissimilar ones. The ABRE standards called for an impartial person familiar with the speaker's voice to be present, whenever possible, when the exemplar was recorded; the examiner did not ask the officers to arrange this. The exemplar was not made with matching equipment or conditions, and the report had no statement of accuracy.
The court also noted that the examiner had moved his opinion over the years: "possible elimination" in 1998, "probable elimination" in 2003, and that a second analyst reached a different conclusion while finding that most phrases lacked sufficient quality for comparison.
Recording conditions change the result of automatic systems too. In a 2018 study by Hughes and colleagues, presented at Interspeech, a semi-automatic system tested on 97 male speakers had an equal error rate of 10.33% when both recordings were studio quality and 25.95% when the questioned recording came over a landline. The reviewer compares the channel, length and speaking style of the known and questioned recordings.
Morrison and Thompson, in a 2017 article in the Columbia Science and Technology Law Review, discuss how forensic voice comparison systems are validated and how contextual bias may influence the examiner, and they review Angleton and the Daubert hearing in United States v. Ahmed (2015). The reviewer asks whether the method was validated and what the examiner knew about the case before comparing.
What should the reviewer check in enhanced audio and transcripts?
Whether enhancement was done on a working copy with the processing documented, whether the original was heard without processing, and whether a transcript is tied to the recording itself.
Enhanced audio is a derivative of the original. SWGDE section 4.3 has the authentication examiner listen critically "without applying any processing," and Appendix A keeps one bit-stream image archived while work is done on another. The reviewer asks for the unprocessed working copy and the list of filters and settings applied to produce the enhanced version, because the facts or data considered under Rule 26(a)(2)(B)(ii) include every step that produced the exhibit.
Rule 901(b)(9) recognizes authentication by "Evidence describing a process or system and showing that it produces an accurate result." An enhancement offered at trial is a product of a process, and the reviewer checks whether the report describes it in enough detail for another examiner to repeat it.
In United States v. Baller, 519 F.2d 463 (4th Cir. 1975), the court of appeals noted that the recordings had been verified for accuracy and that their technical quality had been established by an expert, and that the tapes were played so the jury could make its own aural comparison. A transcript that marks speakers and unclear passages is useful to the reviewer only if it is checked against the recording and the reviewer can hear each disputed passage in the original.
What did counsel tell the opposing examiner?
Communications between counsel and a reporting expert are protected, except for compensation, facts or data counsel provided and the expert considered, and assumptions counsel provided and the expert relied on.
Rule 26(b)(4)(C) protects communications between a party's attorney and a witness required to provide a report, except to the extent that they relate to compensation for the expert's study or testimony, identify facts or data that the attorney provided and the expert considered, or identify assumptions that the attorney provided and the expert relied on. Drafts of any report are protected by Rule 26(b)(4)(B).
In audio cases the exceptions matter. If counsel told the examiner which file to treat as the original, which speaker was the defendant or which words to listen for, the instruction is a fact or an assumption the other side can discover. A transcript prepared by a party and given to the examiner before the comparison is a fact provided by counsel.
Morrison and Thompson's discussion of contextual bias gives the reviewer the reason to ask. An examiner who knows which person is suspected and hears the recording with a party's transcript in hand compares under different conditions than one who receives only the files.
Can the court appoint its own audio expert?
Yes. Rule 706 lets the court, on a party's motion or on its own, appoint an expert. Any party may depose and cross-examine that expert, including the party that called the expert.
Rule 706(a) provides that, on a party's motion or on its own, the court may order the parties to show cause why expert witnesses should not be appointed and may ask the parties to submit nominations. Under Rule 706(b) the appointed expert advises the parties of any findings, may be deposed by any party, may be called to testify by the court or any party, and may be cross-examined by any party, including the party that called the expert.
When the parties' audio experts disagree on the basic facts of the file, such as whether the produced file is the original or whether a discontinuity exists, a motion under Rule 706 is one option. The review gives the court the concrete disagreement to resolve: the hash values, the test results and the points where the two reports part ways.
The decision to appoint, and the choice of expert, belong to the court.
How is the opposing audio examiner deposed and cross-examined?
After the report is served, under Rule 26(b)(4)(A). Rule 705 lets the examining lawyer require the facts or data behind the opinion on cross-examination, and the review supplies the questions.
Rule 26(b)(4)(A) lets a party depose any person identified as an expert whose opinions may be presented at trial, and if Rule 26(a)(2)(B) requires a report, the deposition may be conducted only after the report is provided. Under Rule 26(b)(4)(E), unless manifest injustice would result, the court must require the party seeking discovery to pay the expert a reasonable fee for time spent responding.
Rule 705 lets an expert state an opinion and give the reasons for it without first testifying to the underlying facts or data, "But the expert may be required to disclose those facts or data on cross-examination." Under Rule 703 an expert may base an opinion on facts or data that experts in the field would reasonably rely on, even if those facts are not admissible. Whether audio examiners would rely on a re-encoded copy is a question for the deposition.
The National Institute of Justice course Law 101: Legal Guide for the Forensic Expert lists the lines of cross-examination that apply here: gaps in the résumé, prior writings and testimony directly contrary to the present testimony, authoritative texts in the field, absent measurements, absent personal verification or testing, and reliance on inappropriate second-hand information.
From the review come the specific questions. Which file did you receive and what was its hash? Did you make exemplar recordings on the same model of device? Which tests did you run and which did you skip? What did counsel tell you before you started? Why did your conclusion use absolute language when the SWGDE guide advises against it?
Can the review support a motion to exclude the opposing audio expert?
Yes. Under Rule 702 the party offering the examiner must show that each opinion more likely than not meets the rule's conditions. The review supplies the facts for the motion, and the court decides.
Rule 702 lets a qualified witness testify to an opinion if the proponent shows the court that it is more likely than not that the specialized knowledge will help the trier of fact, that the testimony is based on sufficient facts or data, that it is the product of reliable principles and methods, and that the opinion applies those principles and methods reliably to the facts of the case. The 2023 committee note calls rulings that left the sufficiency of the basis and the application of the method to the jury "an incorrect application of Rules 702 and 104(a)."
The Supreme Court in Daubert, decided June 28, 1993, pointed to testing, peer review and publication, the known or potential error rate, the existence and maintenance of standards controlling the technique's operation, and general acceptance. In Angleton the court found that the potential rate of error of the aural spectrographic method was unknown, that the method had not been generally accepted by the scientific community, and that the examiner had not followed the existing standards. It held the testimony unreliable under Rule 702 and, under Rule 403, unhelpful and confusing.
Baller, decided in 1975, shows the other outcome. The Fourth Circuit found no abuse of discretion in admitting spectrographic voice identification where the examiner was experienced, the recordings were verified, the defense had the names of experts who could analyze the tapes and cross-examined at length, and the jury was told it could disregard the opinion if the method was not sufficiently reliable. The court said that absolute certainty is not required for admissibility.
Daubert also names the remedies short of exclusion: "Vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence." A rebuttal report is contrary evidence and keeps its use if the motion fails. In Joiner the Court held that rulings on expert evidence are reviewed for abuse of discretion, so the record built in the trial court is the record on appeal.
Whether the opposing examiner is excluded, limited or heard in full is decided by the court.
What does IAE do when a party needs a review of an audio expert's report?
An IAE examiner prepares a review of the opposing audio examiner's report for a federal or state case. The document is issued in up to 5 business days after the format and cost are confirmed.
The IAE audio and voice page lists two documents: a report on the client's questions about the speaker, the integrity of the file and the words spoken, and a review of the opposing audio examiner's report. The questions are confirmed with the client before work starts.
According to the page, the review checks which file the examiner worked from, original or copy, whether its hash and metadata are listed among the facts or data considered, as Rule 26(a)(2)(B)(ii) requires, and whether the known samples were enough for the opinion, Rule 702(b). It notes that facts and assumptions the attorney gave the examiner are discoverable under Rule 26(b)(4)(C). On authenticity the examiner checks the file for breaks in background noise, metadata that does not match the device, signs of re-recording and gaps in the chain of copies.
IAE describes a review of an expert report as an assessment of the method, the completeness of the examination, the soundness of the conclusions and the expert's competence. The reviewer answers six questions: whether the method fits the object and the questions and was followed at every stage; whether all objects, materials and questions were examined; whether the conclusions follow from the examination, with no contradictions or inadmissible assumptions; whether the expert's education, specialty and experience match the subject; whether the report meets the formal requirements and procedure of the jurisdiction; and whether the report can be relied on.
Sources
- Federal Rules of Civil Procedure, Rule 26 (Cornell LII)
- Federal Rules of Civil Procedure, Rule 34 (Cornell LII)
- Federal Rules of Civil Procedure, Rule 37 (Cornell LII)
- Federal Rules of Evidence, Rule 702 and committee notes (Cornell LII)
- Federal Rules of Evidence, Rule 703 (Cornell LII)
- Federal Rules of Evidence, Rule 705 (Cornell LII)
- Federal Rules of Evidence, Rule 706 (Cornell LII)
- Federal Rules of Evidence, Rule 901 (Cornell LII)
- Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993) (Cornell LII)
- General Electric Co. v. Joiner, 522 U.S. 136 (1997) (Cornell LII)
- SWGDE Best Practices for Digital Audio Authentication, 15-a-001, version 1.3 (2018)
- United States v. Angleton, 269 F. Supp. 2d 892 (S.D. Tex. 2003) (Justia)
- United States v. Baller, 519 F.2d 463 (4th Cir. 1975) (Justia)
- Koenig B. E., Selected Topics in Forensic Voice Identification, Crime Laboratory Digest 20(4), 1993 (NCJRS)
- Hughes V. et al., The individual and the system: assessing the stability of the output of a semi-automatic forensic voice comparison system, Interspeech 2018
- Morrison G. S., Thompson W. C., Assessing the Admissibility of a New Generation of Forensic Voice Comparison Testimony, Columbia Science and Technology Law Review 18(2), 2017 (NIJ abstract)
- National Institute of Justice, Law 101: Legal Guide for the Forensic Expert, cross-examination
Read also
All articles
ShotSpotter evidence: is it admissible, and how does a reviewer test the other side's acoustic gunshot report?
An acoustic gunshot report tells the jury how many shots were fired, when and where, on the basis of sound picked up by sensors. Courts have admitted
The opposing audio report rests on a compressed file from a messaging app or a phone: what does a reviewer check?
Most recordings in disputes today reach the examiner as compressed files: voice messages, call recordings, memos made on a phone. Compression
Was the opposing audio examiner told what to find, and how does a reviewer check for contextual bias?
An audio examiner who knows that the suspect confessed, or that other evidence points to him, is working with information that the examination doesExpert report review
Ready to discuss your case?
Send the materials: the secretariat will confirm the format, cost, deadline and the specialist within one business day.

