Projects

*

Innovación docente en Análisis del Discurso a través de textos humorísticos continuos y discontinuos

Teaching Innovation Project aimed at promoting students’ motivation, participation and performance improvement

How deepfake is your voice? Understanding the linguistic foundations of deepfakes

Research project funded by the Spanish Ministry of Science and Innovation

Metodologías de aprendizaje activo en la lingüística y sus aplicaciones

Teaching Innovation Project implementing problem-based learning methods to Applied Linguistics (Forensic Linguistics, Clinical Linguistics)

Research on Key Technologies for Forensic Speaker Comparison

International project at the Shanxi Police College (Taiyuan, China)

Perceptual, spectral and prosodic correlates of vocal tract tension

SNSF project aimed at testing whether perceptual ratings for vocal tract tension correlate with prosodic-acoustic measures.

How to deal with voice quality in Forensic Phonetics

IAFPA project aimed at providing a simplified, yet reliable and validated, perceptual protocol that could be used by forensic phoneticians to describe a speaker in terms of his/her voice quality, and/or when required to compare two or more voices using an auditory approach.

Voice and Identity: Source, Filter, Biometric

Arts & Humanities Research Council (AHRC) project

Forensic comparison of Spanish twins and non-twin brothers

IAFPA project aimed at investigating the usefulness of analyzing speech samples of adult male Spanish siblings and twins (identical and fraternal), with a focus on analyzing formant trajectories in vocalic sequences.

Incorporation of forensic analysis techniques as part of an automatic speaker recognition system

University of New South Wales project funded by the Intelligence Advanced Research Projects Activity (IARPA), through the Army Research Laboratory (ARL)

Cualidad Individual de la Voz en la Identificación de Locutor

Research project funded by the Spanish Ministry of Science and Innovation

Recent Publications

Click below to see ALL publications

More Publications

(2024). Profundizando en los deepfakes: ¿Qué hace humana a una voz?. Anuario AC/E 2024 de Cultura Digital (pp. 28-41), Acción Cultural Española (AC/E). ISBN: 978-84-17265-41-0.

Link

(2023). 10th anniversary of Loquens. Loquens, 10 (1-2), e105. https://doi.org/10.3989/loquens.2023.e105.

Link

(2023). ¿Voz real o deepfake? Aportaciones desde la fonética forense. CIFE 2023: Book of Abstracts (p. 176).

Link

(2023). Characterizing rhythm in dysarthric speech using the temporal envelope. In: Radek Skarnitzl & Jan Volín (Eds.), Proceedings of the 20th International Congress of Phonetic Sciences (pp. 3907–3911). Guarant International. ISBN 978-80-908 114-2-3.

PDF

Conference and Invited Talks

More Talks

Speaker identification of artificially cloned voices versus the voices of identical twins: a perceptual test.
Sep 11, 2024
Voice deepfakes and the voices of identical twins: a preliminary acoustic analysis
Sep 11, 2024

Previous positions

 
 
 
 
 
March 2020 – January 2024
Madrid, Spain

Distinguished Researcher Beatriz Galindo

Department of Spanish Language & General Linguistics (UNED)

I taught two undergraduate courses (Applied Linguistics and Discourse Analysis) and two postgraduate courses (Information Technologies in Spanish L2 Teaching and Corpus Linguistics in Spanish L2 Teaching), coordinated two teaching innovation projects, codirected a lifelong learning course on forensic phonetics, and supervised 5 master theses and 5 undergraduate theses.
 
 
 
 
 
August 2018 – February 2020
Geneva, Switzerland

Linguistic consultant & phonetician

DefinedCrowd Corp.

I collaborated in several projects, mainly aimed at improving the quality of the Vocalizer TTS product of Nuance Communications Inc., providing verified phonetic transcriptions of music and news items, researching the pronunciation of new words and correcting transcriptions for words that failed the automatic-system checks and manual quality checks.

Responsibilities included:

  • Phonetic transcription
  • Research on pronunciation
  • Validation of recorded calls
  • Translation EN>ES
  • Sentence verification for AI system training
  • PoS tagging & other NLP tasks

Main skills:

  • Expert knowledge of phonetic alphabets (IPA, SAMPA and X-SAMPA)
  • Ability to familiarize with in-house user guides and language-specific transcription conventions

 
 
 
 
 
May 2018 – February 2020
Taiyuan, China

Research Associate

Department of Criminal Science & Technology, Shanxi Police College

Project: Research on Key Technologies for Forensic Speaker Comparison in Mandarin and English
 
 
 
 
 
May 2015 – September 2017
York, UK

Postdoctoral Fellow

Department of Language & Linguistic Science, University of York

Responsibilities included:

  • Annotation and review of large British English speech corpus
  • Assessment of forensic system performance
  • Voice quality perceptual evaluation of 100 speakers
  • Led the voice quality research line: method reliability and validity
  • Clinical and speech processing tools

 
 
 
 
 
January 2010 – December 2014
Madrid, Spain

PhD Student

Laboratorio de Fonética (CSIC)

Doctor Europaeus - Thesis: “Forensic speaker comparison of Spanish twins and non-twin siblings: A phonetic-acoustic analysis of formant trajectories in vocalic sequences, glottal source parameters and cepstral characteristics”

Achievements: - Largest twin data collection to date in Spanish (54 subjects, 108 hours). - Developed strong managerial and statistical skills. - Self-managed with minimal direction: responsible for all the logistics regarding equipment, participant identification and performing the data collection. - Tested different approaches to speaker ID: acoustic and semi-automatic analyses (ASR system output evaluation).

Main skills: - Strong analytical skills - Attention to detail - Excellent written communication skills - Ability to grasp technical concepts (collaboration with engineers and software developers: co-authors in my main publications) - Hands-on testing of several in-house tools (Biometrics software); monitored debugging process results - Knowledge of curve fitting methods and use of Bayesian data analysis.

 
 
 
 
 
January 2009 – July 2009
Madrid, Spain

Translator

Eurydice (Spanish unit), Ministry of Education, Spain

Translator (EN>ES; FR>ES) of educational publications, and researcher on European and Spanish education.
 
 
 
 
 
October 2007 – April 2008
Valladolid, Spain

Teacher of Spanish as a Foreign Language

La Casa del Español

Teacher of Spanish language, literature and culture in one-to-one classes, mainly to English-speaking adults aged 18-40 of different academic and professional backgrounds.

Skills

Statistics

R

Scripting

Trello

Audio recording and editing

Phonetic transcription

Accomplish­ments

Apr 2024

Acreditación de la ANECA para la figura de Profesora Titular de Universidad

Agencia Nacional de Evaluación de la Calidad y Acreditación

Favourable lecturer report issued by the National Agency for Quality Assessment and Accreditation of Spain https://www.aneca.es/acreditaciones-conseguidas

Challenging Forensic Science: How Science Should Speak to Court

Coursera

School of Criminal Justice, University of Lausanne (Switzerland)
May 2018

Acreditación de la ANECA para la figura de Profesor Contratado Doctor y Profesor de Universidad Privada

Agencia Nacional de Evaluación de la Calidad y Acreditación

Favourable lecturer report issued by the National Agency for Quality Assessment and Accreditation of Spain
May 2018

Acreditación de la ANECA para la figura de Profesor Ayudante Doctor

Agencia Nacional de Evaluación de la Calidad y Acreditación

Favourable lecturer report issued by the National Agency for Quality Assessment and Accreditation of Spain

Contact

  • eugenia.sansegundo@csic.es
  • Centro de Ciencias Humanas y Sociales (CSIC),
    Instituto de Lengua, Literatura y Antropología,
    Despacho 1E16.
    Calle Albasanz 26-28, 28037 Madrid