The Geneva Minimalistic Acoustic Parameter Set (GeMAPS) for Voice Research and Affective Computing

By Florian Eyben, Klaus R. Scherer, Björn Schuller, J. Sundberg, Elisabeth André, Carlos Busso, Laurence Y. Devillers, Julien Epps, Petri Laukka, Shrikanth S. Narayanan and 1 more. IEEE Transactions on Affective Computing, 2016

1,837 citations recorded by the source corpus.

Work on voice sciences over recent decades has led to a proliferation of acoustic parameters that are used quite selectively and are not always extracted in a similar fashion. With many independent teams working in different research areas, shared standards become an essential safeguard to ensure compliance with state-of-the-art methods allowing appropriate comparison of results across studies and potential integration and combination of extraction and recognition systems. In this paper we propose a basic standard acoustic parameter set for various areas of automatic voice analysis, such as paralinguistic or clinical speech analysis. In contrast to a large brute-force parameter set, we present a minimalistic set of voice parameters here. These were selected based on a) their potential to index affective physiological changes in voice production, b) their proven value in former studies as well as their automatic extractability, and c) their theoretical significance. The set is intended to provide a common baseline for evaluation of future research and eliminate differences caused by varying parameter sets or even different implementations of the same parameters. Our implementation is publicly available with the openSMILE toolkit. Comparative evaluations of the proposed feature set and large baseline feature sets of INTERSPEECH challenges show a high performance of the proposed set in relation to its size.

Subfields: Artificial Intelligence, Signal Processing

View Paper: https://www.semanticscholar.org/paper/10f992779c2601af8a33f53c813cb6342791873f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Signal Processing

Citations in the corpus

AVEC 2016: Depression, Mood, and Emotion Recognition Workshop and Challenge (2016, methodology): “Some recommendations for the definition of a minimalistic acoustic standard parameter set have been recently investigated, and have led to the Geneva Minimalistic Acoustic Parameter Set (GeMAPS), and to an extended version (eGeMAPS) [10], which is used here as baseline.”
Speech Emotion Recognition Using Deep Convolutional Neural Network and Discriminant Temporal Pyramid Matching (2018, methodology): “Note that some pervious works [5], [62] also employ Unweighted Averaged recall (UAR), which is used Fig.”
Alzheimer's Dementia Recognition through Spontaneous Speech: The ADReSS Challenge (2020, background): “eGeMAPS: The eGeMAPS [14] feature set resulted from an attempt to reduce the somewhat unwieldy feature sets above to a basic set of acoustic features based on their potential to detect physiological changes in voice production, as well as theoretical significance and proven usefulness in previous st”
Multimodal Continuous Emotion Recognition with Data Augmentation Using Recurrent Neural Networks (2018, methodology): “This challenge adopts eGeMAPS [12] as the baseline audio features.”
End-to-End Multimodal Emotion Recognition Using Deep Neural Networks (2017, background): “One may note, however, that the eGeMAPS [46] feature set provides close performance on valence, which is much more difficult to predict from speech compared to arousal.”
Building Naturalistic Emotionally Balanced Speech Corpus by Retrieving Emotional Speech from Existing Podcast Recordings (2019, methodology): “ture set for this experiment is the extended version of the Geneva minimalistic acoustic parameter set (eGeMAPS) [62], extracted with OpenSMILE [63].”
Automated assessment of psychiatric disorders using speech: A systematic review (2019, background): “For further discussion, see the Geneva Minimalistic Acoustic Parameter Set (GeMAPS) 57 and Section 4.3.3”
Emotion Recognition from Speech Using Wav2vec 2.0 Embeddings (2021, methodology): “As baseline features, we also calculated magnitude spectrograms with a hanning window of 25 ms and a hop size of 10 ms, and eGeMAPS [32] low level descriptors (LLD), which are commonly used in the emotion recognition literature.”
AVEC 2019 Workshop and Challenge: State-of-Mind, Detecting Depression with AI, and Cross-Cultural Affect Recognition (2019, methodology): “As audio features, we compute the extended GenevaMinimalistic Acoustic Parameter Set (eGeMAPS) [19], which contains 88 measures covering the aforementioned acoustic dimensions, and used here as baseline.”
Investigating Speech Features for Continuous Turn-Taking Prediction Using LSTMs (2018, positive): “As acoustic features, we use the low-level descriptors from the eGeMAPs [12] feature set extracted with the OpenSmile toolkit [13].”
Introducing AmuS: The Amused Speech Database (2017, methodology): “Recognition has been largely based on feature extraction for later classification [15].”
An Overview of Affective Speech Synthesis and Conversion in the Deep Learning Era (2022, background): “This is also encapsulated in the design of SER feature sets that incorporate voice quality measures as markers related to emotion [148].”
Adieu features? End-to-end speech emotion recognition using a deep convolutional recurrent network (2016, methodology): “Some of these features, such as the mean of the fundamental frequency (F0), mean speech intensity, loudness, as well as pitch range [23], should thus be captured by our model.”
Analysis and Assessment of Controllability of an Expressive Deep Learning-based TTS system (2021, background): “These features are defined in [3] as follows:”