Limitations – JP French International

The amount of testing we can do on forensic recordings is sometimes limited. Where this is so, we may still be able to come to a conclusion, but it could be a weak one – or in some cases, the outcome may be inconclusive.

Things that can limit our testing include:

Brevity

Recordings that contain little speech are more likely to be insufficient for a full analysis.

Poor Technical Quality

Poor technical quality can affect the types of analyses that are possible.

Noise

Noise in recordings can ‘mask’ the speech making it difficult to analyse fully.

Speaking style

Some styles of speech can be less suitable for analysis, such as reading aloud or talking to small children or pets.

Speaking level

Some levels of speech can be less suitable for analysis, such as shouting or whispering.

Distance

Speech that is recorded far away from the microphone can be less clear and cannot always be analysed.

Voice disguise

Disguised voices cannot always undergo the same analyses as non-disguised speech.

Mismatch

If comparison recordings are mismatched by any of the aforementioned limitations, they could produce a weaker conclusion.

How we present our conclusions

Service-specific limitations

Speaker Comparison

Speaker comparison is the process of comparing the voice and speech patterns of a known suspect with an unknown voice on a recording or recordings.

We examine and compare the voices using a number of phonetic and acoustic parameters. If the recording(s) are affected by one or more of the limitations, the number of parameters that can be measured are reduced and the voice profiles may be partial.

These comparisons have the potential to produce a conclusion but there is a risk that the outcomes are low strength or inconclusive. We will make you aware of any initial limitations at the screening stage and make you aware of any limitations that become apparent during analysis in the written report. Often some limitations are not obvious until the analysis has been completed.

Enhancement

Enhancement is the digital processing of recordings to reduce noise, change levels or emphasise parts of the audio signal. The purpose is to improve the intelligibility of speech in recordings with limitations.

Whilst some noise contamination can be improved through enhancement, there are limitations to what is possible. For example, it is not possible to ‘restore’ speech when it is not clearly recorded at the time of the event, nor is it possible to separate foreground sound from background sound when the two overlap. We also cannot make whispered or shouted speech into normal speaking level speech.

We use professional-grade software and hardware to produce the best possible results; sometimes speech cannot be made clearer despite enhancement processing.

Transcription

Forensic transcription is the process of writing down the speech sounds within a poor quality or covert recording. Often we enhance the recording before we produce an evidential or investigative transcript.

The benefits of a transcript from J P French Associates are that we provide an independent, expert transcript that is subject to a bias management policy, we use high quality, professional-grade equipment in a quiet, laboratory environment and all transcripts undergo a rigorous checking procedure by a secondary expert.

Limitations can result in sections of unintelligible speech. We only include interpretations that we are confident in and can stand up in court to defend. We do not include speech that we cannot reliably interpret, even if our clients have their own interpretations about the speech content.