Tuning the performance of automatic speaker recognition in different conditions

Skarnitzl, Radek; Asiaee, Maral; Nourbakhsh, Mandana

doi:10.1558/ijsll.39778

Cited by 2 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fabien and Motlicek [31] investigated the performance of x‐vector models in forensic scenarios, but with acted speech, and the study did not focus on the effect of multilingualism (although the dataset was multilingual). The study by Skarnitzl and colleagues [32] is specifically related to forensic research and evaluated multilingual scenarios but uses an earlier version of the VOCALIZE [33] system based on the obsolete i‐vector.…”

Section: Introductionmentioning

confidence: 99%

Effects of language mismatch in automatic forensic voice comparison using deep learning embeddings

Schiferl

Fejes

2023

Journal of Forensic Sciences

View full text Add to dashboard Cite

In forensic voice comparison, deep learning has become widely popular recently. It is mainly used to learn speaker representations, called embeddings or embedding vectors. Speaker embeddings are often trained using corpora mostly containing widely spoken languages. Thus, language dependency is an important factor in automatic forensic voice comparison, especially when the target language is linguistically very different from that the model is trained on. In the case of a low‐resource language, developing a corpus for forensic purposes containing enough speakers to train deep learning models is costly. This study aims to investigate whether a model pre‐trained on multilingual (mostly English) corpus can be used on a target low‐resource language (here, Hungarian), not represented by the model. Often multiple samples are not available from the offender (unknown speaker). Samples are therefore compared pairwise with and without speaker enrollment for suspect (known) speakers. Two corpora are used that were developed especially for forensic purposes and a third that is meant for traditional speaker verification. Speaker embedding vectors are extracted by the x‐vector and ECAPA‐TDNN techniques. Speaker verification was evaluated in the likelihood‐ratio framework. A comparison is made between the language combinations (modeling, LR calibration, and evaluation). The results were evaluated by Cllrmin and EER metrics. It was found that the model pre‐trained on a different language but on a corpus with a significant number of speakers can be used on samples with language mismatch. Sample duration and speaking style also seem to affect the performance.

show abstract

Section: Introductionmentioning

confidence: 99%

Effects of language mismatch in automatic forensic voice comparison using deep learning embeddings

Schiferl

Fejes

2023

Journal of Forensic Sciences

View full text Add to dashboard Cite

show abstract

“…Apart from the three comparisons listed above, we conducted several partial comparisons to examine the effect of "tuning" (see Skarnitzl et al, 2019) using condition adaptation. Condition adaptation optimizes the ASR system to new conditions, specific to the given recordings, by adapting the LDA and PLDA models.…”

Section: Automatic Speaker Recognition Proceduresmentioning

confidence: 99%

The impact of mismatched recordings on an automatic-speaker-recognition system and human listeners

Nechanský¹,

Bořil²,

Houzar³

et al. 2023

Auc Philologica

Self Cite

View full text Add to dashboard Cite

The so-called ‘mismatch’ is a factor which experts in the forensic voice comparison field encounter regularly. Therefore, we decided to explore to what extent the automatic-speaker-recognition system’s and the earwitness’ ability to identify speakers is influenced when recordings are acquired in different languages and at different times. 100 voices in a database of 300 recordings (100 speakers recorded in three mutually mismatched sessions) were compared with an automatic-speaker-recognition software VOCALISE based on i-vectors and x-vectors, and by 39 respondents in simulated voice parades. Both the automatic and perceptual approach seem to have yielded similar results in that the less complex the mismatch type, the more successful the identification. The results point to the superiority of the x-vector approach, and also to varying identification abilities of listeners.

show abstract

Tuning the performance of automatic speaker recognition in different conditions

Cited by 2 publications

References 27 publications

Effects of language mismatch in automatic forensic voice comparison using deep learning embeddings

Effects of language mismatch in automatic forensic voice comparison using deep learning embeddings

The impact of mismatched recordings on an automatic-speaker-recognition system and human listeners

Contact Info

Product

Resources

About