Fusing linguistic and acoustic information for automated forensic speaker comparison.

Verifying the speaker of a speech fragment can be crucial in attributing a crime to a suspect. The question can be addressed given disputed and reference speech material, adopting the recommended and scientifically accepted likelihood ratio framework for reporting evidential strength in court. In forensic practice, usually, auditory and acoustic analyses are performed to carry out such a verification task considering a diversity of features, such as language competence, pronunciation, or other linguistic features. Automated speaker comparison systems can also be used alongside those manual analyses. State-of-the-art automatic speaker comparison systems are based on deep neural networks that take acoustic features as input. Additional information, though, may be obtained from linguistic analysis. In this paper, we aim to answer if, when and how modern acoustic-based systems can be complemented by an authorship technique based on frequent words, within the likelihood ratio framework. We consider three different approaches to derive a combined likelihood ratio: using a support vector machine algorithm, fitting bivariate normal distributions, and passing the score of the acoustic system as additional input to the frequent-word analysis. We apply our method to the forensically relevant dataset FRIDA and the FISHER corpus, and we explore under which conditions fusion is valuable. We evaluate our results in terms of log likelihood ratio cost (Cllr) and equal error rate (EER). We show that fusion can be beneficial, especially in the case of intercepted phone calls with noise in the background.

Discrimination of coal geographical origins through HS-GC-IMS assisted with machine learning algorithms in larceny case.

The process of globalization and industrialization has resulted in a rise in the theft of coal and other related products, thereby becoming a focal point for forensic science. This situation has engendered an escalated demand for effective detection and monitoring technologies. The precise identification of coal trace evidence presents a challenge with current methods, owing to its minute quantity, fine texture, and intricate composition. In this study, we integrated machine learning with the identification of volatiles to accurately differentiate coal geographical origins through the application of headspace-gas chromatography-ion mobility spectrometry (HS-GC-IMS). The topographic distribution of volatiles in coals was visually depicted to elucidate the subtle distinctions through spectra and fingerprint analysis. Additionally, four supervised machine learning algorithms were developed to quantitatively predict the geographical origins of natural coals utilizing the HS-GC-IMS dataset, and these were subsequently compared with unsupervised models. Remarkable volatile compounds were identified through the quantitative analysis and optimal Random Forest model, which offered a rapid readout and achieved an average accuracy of 100 % in coal identification. Our findings indicate that the integration of HS-GC-IMS and machine learning is anticipated to enhance the efficiency and accuracy of coal geographical traceability, thereby providing a foundation for litigation and trials.

Use of innovative sample dryer for effective sample drying in forensic investigations: An experimental study.

BACK GROUND: The sample collection and preservation before transportation to a Forensic Science Laboratory in a medicolegal case in India is a crucial step in establishing a link between accused, victim, crime scene and the weapon. This paper discusses pioneering advancement in health technology-a state-of-the-art sample dryer developed to preserve biological samples during Forensic investigations. This unique design has been officially registered with the design office in India as a copy right. AIMS AND OBJECTIVES: The primary goal of the sample dryer is to augment the preservation of biological evidence by removing moisture content through air drying before packaging, sealing and its transportation to a Forensic Science Laboratory. MATERIAL AND METHODS: The process of air drying of biological samples by variety of experiments conducted in sample's dryer designed for the purpose which employs a range of air movement techniques including horizontal, vertical, and circular laminar flows to swiftly extract moisture from the samples. The drying duration for each experimental sample was documented upon manual confirmation of complete drying. RESULTS: The results were extraordinary, showcasing that the sample's dryer reduced the drying time drastically by up to 27 times lesser in comparison to conventional methods for identical samples. The diverse airflow patterns generated by the sample dryer unequivocally demonstrated it's utility in achieving superior sample preservation through efficient air drying. CONCLUSION: Given the exceptional outcomes observed in the experimental phase, the authors strongly advocate for the widespread adoption of this innovative sample's dryer equipment to ensure effective air drying of biological samples and their efficient sample preservation to enable successful forensic analysis as expected.

Benchmarking automation-aided performance in a forensic face matching task.

Carragher and Hancock (2023) investigated how individuals performed in a one-to-one face matching task when assisted by an Automated Facial Recognition System (AFRS). Across five pre-registered experiments they found evidence of suboptimal aided performance, with AFRS-assisted individuals consistently failing to reach the level of performance the AFRS achieved alone. The current study reanalyses these data (Carragher and Hancock, 2023), to benchmark automation-aided performance against a series of statistical models of collaborative decision making, spanning a range of efficiency levels. Analyses using a Bayesian hierarchical signal detection model revealed that collaborative performance was highly inefficient, falling closest to the most suboptimal models of automation dependence tested. This pattern of results generalises previous reports of suboptimal human-automation interaction across a range of visual search, target detection, sensory discrimination, and numeric estimation decision-making tasks. The current study is the first to provide benchmarks of automation-aided performance in the one-to-one face matching task.

Do long-term acoustic-phonetic features and mel-frequency cepstral coefficients provide complementary speaker-specific information for forensic voice comparison?

A growing number of studies in forensic voice comparison have explored how elements of phonetic analysis and automatic speaker recognition systems may be integrated for optimal speaker discrimination performance. However, few studies have investigated the evidential value of long-term speech features using forensically-relevant speech data. This paper reports an empirical validation study that assesses the evidential strength of the following long-term features: fundamental frequency (F0), formant distributions, laryngeal voice quality, mel-frequency cepstral coefficients (MFCCs), and combinations thereof. Non-contemporaneous recordings with speech style mismatch from 75 male Australian English speakers were analyzed. Results show that 1) MFCCs outperform long-term acoustic phonetic features; 2) source and filter features do not provide considerably complementary speaker-specific information; and 3) the addition of long-term phonetic features to an MFCCs-based system does not lead to meaningful improvement in system performance. Implications for the complementarity of phonetic analysis and automatic speaker recognition systems are discussed.

Longitudinal Analysis of the Microbial Community on the Surface of Bloodstains Under Different Environmental Conditions in Areas with Minimal Human Interference.

The association between human metabolites and the environmental microbiome has primarily been investigated in relation to disease. In this study, the associations between environmental conditions and microbial communities on the surface of bloodstains were analyzed from a forensic science approach. The composition of microbial communities can be affected by numerous variables. After exposing bloodstains to two different environments with limited airflow and human interference, the microbial communities of the bloodstain surfaces were subjected to longitudinal analysis. Various microbes showed increasing or decreasing trends at the phylum and species level. The microbes identified in this study are usually found in soil, freshwater, and seawater and are known to exhibit unique properties, such as sporulation. Longitudinal variation in temperature and humidity were associated with various changes and correlations with the blood surface microbial community. Understanding these changes could introduce a new perspective to forensic science and could be used to develop a forensic tool used at crime scenes to analyze blood stains in more detail.

Research Progress of Using Canine Olfactory Search for Human Remains.

In the process of murder investigation, it is of great significance to find the discarded and buried human remains accurately. The main methods of searching for human remains include human visual search, aerial detection, geophysical technology, remote imaging technology and canine olfactory search technique. Canine olfactory search for human remains is a recognized time-effective and non-invasive search method, making dogs the most valuable search tool in forensic investigation. By systematically reviewing and summarizing relevant literature, and based on the theory of volatile organic compound produced by the decomposition of human remains, this paper explores the basic principle of the canine olfactory search technique for human remains. This paper also reviews the application of training canine search technique for human remains in forensic investigation by using human blood, tissue, cadaver putrefying fluid and odor substitutes as sniffing sources. The application prospect of canine olfactory search for human remains was prospected from the perspectives of detection of volatile organic compound during cadaver decay, development of odor substitutes and adsorption devices, and technology tactics used in canine training and use, to provide references for the relevant research of canine olfactory search for human remains in China.

Near-infrared excitation Raman spectroscopy of colored fabric contaminated with body fluids.

Confirmatory identification of dyes in the physical pieces of evidence, such as hair and fabric, is critically important in forensics. This information can be used to demonstrate the link between a person of interest and a crime scene. High performance liquid chromatography is broadly used for dye analysis. However, this technique is destructive and laborious. This problem can be overcome by near-Infrared excitation Raman spectroscopy (NIeRS), non-invasive and non-destructive technique that can be used to determine chemical structure of highly fluorescent dyes. Analyzed fabric materials often possess body fluid stains, which may obscure the accuracy of NIeRS-based identification of dyes. In this study, we investigate the extent to which fabric contamination with body fluids can alter the accuracy of NIeRS. Our results showed that NIeRS coupled with partial-least squared discriminant analysis (PLS-DA) enabled on average 97.6% accurate identification of dyes on fabric contaminated with dry blood, urine and semen. We also found that NIeRS could be used to identify blood, urine and semen on such fabric with 99.4% accuracy. Furthermore, NIeRS could be used to differentiate between wet and dry blood, as well as reveal the presence of blood on washed fabric. These results indicate that NIeRS coupled with PLS-DA could be used as a robust and reliable analytical approach in forensic analysis of fabric.

Drying the mystery: a novel electronic sensor to quantify soft-tissue desiccation and natural mummification for forensic taphonomy.

This study investigates the desiccation process of soft-tissue in South Africa, analyzing its interaction with environmental parameters and its implications for estimating the post-mortem interval. Through the examination of four decomposing porcine bodies across two summer seasons and one winter season, the research quantifies desiccation patterns using custom-designed and constructed printed circuit boards to measure the moisture content of body tissue over time. Generalized additive models were used to determine the environmental forces driving desiccation. Tissue resistivity was tested against the environmental predictor variables to determine the amount of variation they account for, and predicted values of the region-specific tissue resistivity variables were measured for each decomposing body. Results reveal distinct desiccation trajectories between summer and winter, with summer conditions conducive to precocious natural mummification. Environmental factors, particularly temperature and solar radiation, emerge as significant drivers of desiccation. This study represents the first quantitative analysis of deep tissue desiccation internationally, but also the first quantitative assessment of desiccation and natural precocious mummification in the Western Cape, South Africa. The exploration of desiccation as a potential indicator for estimating PMI opens new avenues for research and the integration of innovative methodologies and technologies promises to revolutionize forensic taphonomy practices.

The invisible evidence: Digital forensics as key to solving crimes in the digital age.

Digital transformation rapidly changes how we live our lives in the post pandemic world. Unfortunately, digital technology is not limited to law abiding organisations and citizens. Criminal organisations and individuals are quick to identify new opportunities with new technologies, and digital transformation is dramatically changing the character of crimes, terror, and other threats. The fast emergence of new crimes is facilitated by possibilities brought by disruptive technologies such as AI, Internet of Things, drones, and cryptocurrencies that can be disastrous tools in the hands of criminals. Consequently, our society needs far better capacity to prevent and investigate criminal acts to protect organisations and citizens. This brings an urgent need to proactively reform digital forensics to significantly increase our capability to meet the strain on society brought by crimes evolving in the digital transformation era. The future of forensic science is already here, characterized by a mix of opportunities and challenges. It is essential to make it harder to effectively use digital technologies for criminal activities, while leveraging the possibilities of digital technologies by those affected, law enforcement agencies, business and organisations. As digital technologies continue to evolve, we need to stay up to date with the latest developments to effectively investigate and prosecute crimes in the digital age. There is an increased reliance on digital evidence, and the amount of heterogeneous digital evidence in criminal cases keep increasing. The forensic science techniques thus become more sophisticated and play an increasingly important role. However, the scientific area is extremely broad, and beyond the capability of most forensic science labs to keep up with the technology forefront development speed. Besides an urgent need to bring up the subject to the political arena, examples of how we can meet the challenges are discussed such as by extending our cooperation, encourage and facilitate cooperation for training and education to handle the extremely broad and rapid development, working out methods for explaining and visualising evidence for the treatment and legal values of digital evidence in prosecution, and cooperation between product developers and crime investigators for swift innovation of digital forensics tools and methodologies for quickly emerging threats. This paper will highlight specific examples where modern digital techniques are used to solve crimes in the physical world as well as crimes committed in the digital domain and discuss how "good AI" can be used to fight "evil AI" and finally touch on the sensitive balance between the increased power of the new digital forensic tools and private integrity.

How Ontologies Have Supported Digital Forensics: Review and Recommendations.

The evolution of digital media has increased the number of crimes committed using digital equipment. This has led to the evolution of the computer forensics area to digital forensics (DF). Such an area aims to analyze information through its main phases of identification, collection, organization, and presentation (reporting). As this area has evolved, many techniques have been developed, mainly focusing on the formalization of terminologies and concepts for providing a common vocabulary comprehension. This has demanded efforts on several initiatives, such as the definition of ontologies, which are a means to identify the main concepts of a given area. Hence, the existing literature provides several ontologies developed for supporting the DF area. Therefore, to identify and analyze the existing ontologies for DF, this paper presents a systematic literature review (SLR) in which primary studies in the literature are studied. This SLR resulted in the identification of ontology building methodologies, ontology types, feasibility points, evaluation/assessment methods, and DF phases and subareas ontologies have supported. These results were based on the analysis of 29 ontologies that aided in answering six research questions. Another contribution of this paper is a set of recommendations on further ontology-based support of DF investigation, which can guide researchers and practitioners in covering existing research gaps.

Similarity, analogy, compatibility, identification: personal identification in forensic practice.

Background: Personal identification is a forensic procedure of recognizing a subject, whether corpse or living, based on a set of parameters and evidence. Spreading of video surveillance cameras has changed forensic application in identification. Indeed, there has been a shift from the evaluation of universally recognized anthropometric parameters to the spasmodic search for individual traits and/or characteristics that could be safely attributed to a single subject. Case Series: Authors analyzed two different cases of personal identification with the analysis and processing of images extracted from video surveillance systems. The first case involved two individuals suspected of throwing an ordnance into the relevant space of a public building. The second case involved a subject suspected of committing a robbery against a commercial establishment. Conclusion: In the field of personal identification focused on images processing extracted from video surveillance systems, as in the cases presented by the authors, there are lot of critical issues and limitations, which could undermine data. Correct personal identification can be achieved by comparing the clearly visible and objectively recognizable somatophysical and physiognomic characteristics of the offender with those of the suspects. So how compatible are two subjects compared? Compatibility and identification are two very different con-cepts. It is not necessarily the case that full compatibility is equivalent to personal identification. It is not necessarily the case that the absence of morphological similarity is equivalent to identification exclusion.

The effects of F-500 Encapsulator Agent on canine alerts and laboratory analysis using passive headspace concentration and gas chromatography-mass spectrometry.

F-500 Encapsulator Agent (EA) is a fire suppression agent that is an alternative to traditional firefighting foams. It is marketed as having the capability to act on all four parts of the fire tetrahedron as well as being environmentally friendly and non-toxic. An internal survey of the use of F-500 EA by fire departments encountered by the Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) ignitable liquid detection canine (ILDC) handlers in 2022 showed that this product is not yet in widespread use across the country, but where it has been implemented, it is frequently utilized on a variety of types of fires. Additional agencies are researching the product to determine if it should be adopted. As this product appears to be growing in popularity, it is important to understand whether the use of the product would affect a canine's ability to detect ignitable liquids or a forensic laboratory's ability to identify the presence of an ignitable liquid. Burned wood and burned carpet, two commonly encountered substrates, were spiked with gasoline or a heavy petroleum distillate (HPD) and F-500 EA was applied. At various time intervals, ILDC teams surveyed the samples and laboratory analysis was conducted. Results showed that the presence of F-500 EA can negatively affect canine alerts and the laboratory's ability to identify ignitable liquids.

Skin locations inference and body fluid identification from skin microbial patterns for forensic applications.

Given that microbiological analysis can be an alternative method that overcomes the shortcomings of traditional forensic technology, and skin samples may be the most common source of cases, the analysis of skin microbiome was investigated in this study. High-throughput sequencing targeting the V3-V4 region of 16S rRNA gene was performed to reveal the skin microbiome of healthy individuals in Guangdong Han. The bacterial diversity of the palm, navel, groin and plantar of the same individual was analyzed. The overall classification based on 16S rRNA gene amplicons revealed that the microbial composition of skin samples from different anatomical parts was different, and the dominant bacterial genus of the navel, plantar, groin and palm skin were dominated by Cutibacterium, Staphylococcus, Corynebacterium and Staphylococcus, respectively. PCoA analysis showed that the skin at these four anatomical locations could only be grouped into three clusters. A predictive model based on random forest algorithm showed the potential to accurately distinguish these four anatomical locations, which indicated that specific bacteria with low abundance were the key taxa. In addition, the skin microbiome in this study is significantly different from the dominant microbiome in saliva and vaginal secretions identified in our previous study, and can be distinguished from these two tissue fluids. In conclusion, the present findings on the community and microbial structure details of the human skin may reveal its potential application value in assessing the location of skin samples and the type of body fluids in forensic medicine.

A comprehensive review on application of atomic force microscopy in Forensic science.

The primary objective of forensic investigation of a case is to recognize, identify, locate, and examine the evidence. Microscopy is a technique that provides crucial information for resolving a case or advancing the investigation process by analyzing the evidence obtained from a crime scene. It is often used in conjunction with suitable analytical techniques. Various microscopes are employed; scanning probe microscopes are available in diverse forensic analyses and studies. Among these, the atomic force microscope (AFM) is the most commonly used scanning probe technology, offering a unique morphological and physico-chemical perspective for analyzing multiple pieces of evidence in forensic investigations. Notably, it is a non-destructive technique capable of operating in liquid or air without complex sample preparation. The article delves into a detailed exploration of the applications of AFM in the realms of nanomechanical forensics and nanoscale characterization of forensically significant samples.

Kafka's beautiful eyes: Forensic intelligence utilisation of phenotypic information.

Franz Kafka had beautiful eyes. So striking, that many of the famous author's friends and peers commented on them - but quite variously ('dark', 'brown', 'grey' & 'blue'). Eye colour as perceived by an observer is subjective, being influenced by physiological, environmental, and even sociocultural factors. In a policing context, this does not mean that trait information such as eye colour is not valuable (far from it), but that it must be managed carefully. The Australian Federal Police has recently implemented a forensic DNA phenotyping (FDP, aka. physical trait prediction or PTP) capability, utilising massively parallel sequencing DNA technology to predict an individual's eye colour, biogeographical ancestry and sex from a crime scene sample. This information alone is not itself 'intelligence', but can be used to generate intelligence through holistic analyses undertaken within a transdisciplinary, all-source forensic intelligence (FORINT) framework. FORINT outputs posit abductive propositions typically at the activity/offence level, to provide insight and influence decision making. However, the use of predicted traits requires that they are compared to something; all Australian police databases include fields for physical traits, but no uniform standard is applied across all agencies. Moreover, collection is inconsistent and no automated systems are in place to capture such data systematically. Consider the 'Kafka problem': his peers gave multiply divergent descriptions of his eyes. If a Biology unit had predicted the eye colour of an 'unidentified author' using DNA - how would Kafka be confidently nominated as the contributor? We posit three maxims for law enforcement: (1) To expand the operational utility of forensic science in line with police demands, forensic science should operationalise FDP (e.g. operationally to rank a list of persons of interest, focus lines of enquiry in serious & organised crime, or assist with human remains identification). (2) Such advanced biological techniques are best delivered through an all-source FORINT framework, to maximise opportunities and minimise risk. (3) One cannot pursue techno-scientific advancements in isolation; it is also necessary to influence the operational posture for their implementation. In this paper we explore these issues and provide recommendations relating to (a) police practices, (b) image capture systems, and (c) research opportunities. Phenotypic trait prediction has great potential and can be operationalised effectively through a rigorous FORINT framework. However, there is (continual) work to be done to enhance the operational capabilities that are complementary to - but necessary for - effective forensic science contribution to investigations.

Optimization and standardization of procedures in Forensic Identification: A methodology for description and coding of tattoos in Mexico.

Tattoos serve as a valuable tool for identification. In the forensic context, it is vital to establish a systematic approach for documenting tattoo-related information to facilitate efficient and fast comparisons, especially in postmortem cases. Despite some countries failing to recognize the potential significance of tattoos, this study presents a methodological framework for gathering comprehensive data on this form of body modification. This article presents the results of an investigation made in Mexico during 2019-2022. The proposed methodology introduces a systematic and distinct classification system tailored to the country in which it will be implemented. The proposal is accompanied by applying the methodology in a Forensic Medical Service (SEMEFO) in Mexico for a week to test its effectiveness and speed under high workloads and stressful conditions. The novelty of this article lies in emphasizing the need for established, replicable, and homologous methodologies for tattoo codification. Additionally, it presents a more in-depth codification, where the details of the tattoos to be classified are thoroughly analyzed.

The forensic potential of environmental DNA (eDNA) in freshwater wildlife crime investigations: From research to application.

Environmental DNA (eDNA) is widely used in biodiversity, conservation, and ecological studies but despite its successes, similar approaches have not yet been regularly applied to assist in wildlife crime investigations. The purpose of this paper is to review current eDNA methods and assess their potential forensic application in freshwater environments considering collection, transport and persistence, analysis, and interpretation, while identifying additional research required to present eDNA evidence in court. An extensive review of the literature suggests that commonly used collection methods can be easily adapted for forensic frameworks providing they address the appropriate investigative questions and take into consideration the uniqueness of the target species, its habitat, and the requirements of the end user. The use of eDNA methods to inform conservationists, monitor biodiversity and impacts of climate change, and detect invasive species and pathogens shows confidence within the scientific community, making the acceptance of these methods by the criminal justice system highly possible. To contextualise the potential application of eDNA on forensic investigations, two test cases are explored involving i) species detection and ii) species localisation. Recommendations for future work within the forensic eDNA discipline include development of suitable standardised collection methods, considered collection strategies, forensically validated assays and publication of procedures and empirical research studies to support implementation within the legal system.