Recent developments in openSMILE, the munich open-source multimedia feature extractor

By Florian Eyben, Felix Weninger, Florian Groß, Björn Schuller. ACM Multimedia, 2013

1,330 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

View Paper: https://www.semanticscholar.org/paper/bd1c73f96ca14ca3bf09d6bec0cd6f66b5bc71a8

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

A Survey on perceived speaker traits: Personality, likability, pathology, and the first challenge (2015, background): “Correspondingly, computing approaches deal with personality under two main perspectives, namely automatic personality recognition (APR), aimed at mapping behaviour into self-assessed traits, and automatic personality perception (APP), aimed at mapping behaviour into traits assigned by others. Pianes”
Controlling Emotion Strength with Relative Attribute for End-to-End Speech Synthesis (2019, positive): “Specifically we use openSMILE tool [27] to extract low-level descriptor contours (e.”
The INTERSPEECH 2020 Computational Paralinguistics Challenge: Elderly Emotion, Breathing & Masks (2020, background): “It contains 6 373 static features resulting from the computation of functionals (statistics) over low-level descriptor (LLD) contours [22, 21].”
AST: Audio Spectrogram Transformer (2021, background): “E[1] E[2] E[3] E[4] E[5] E[6] E[7] E[8] P[1] P[2] P[3] P[4] P[5] P[6] P[7] P[8] Linear Output”
Emotional Voice Conversion: Theory, Databases and ESD (2021, methodology): “In emotion recognition, we also rely on similar acoustic features (Schuller and Schuller, 2020; Latif et al., 2020), for example, expert-crafted features such as those extracted with openSMILE (Eyben et al., 2010, 2013), and learnt acoustic features from the spectrum (Ghosh et al., 2016).”
Predicting Visual Features From Text for Image and Video Caption Retrieval (2017, methodology): “By the time of TrecVid submission, we used GoogLeNet-shufﬂe as the visual feature, a 1,024-dim bag of MFCC as the audio feature, and word2vec for sentence vectorization.”
madmom: A New Python Audio and Music Signal Processing Library (2016, background): “Some libraries concentrate on low-level feature extraction from audio signals, such as Marsyas [18], YAAFE [14] and openSMILE [9].”
AVEC 2019 Workshop and Challenge: State-of-Mind, Detecting Depression with AI, and Cross-Cultural Affect Recognition (2019, methodology): “In addition, MFCCs 1-13, including their 1 st - and 2 nd -order derivatives (deltas and double-deltas) are computed as a set of acoustic LLDs, using the openSMILE 2 [20] toolkit.”
The Geneva Minimalistic Acoustic Parameter Set (GeMAPS) for Voice Research and Affective Computing (2016, positive): “All the parameters are extracted with the open-source toolkit openSMILE [59].”
The INTERSPEECH 2016 Computational Paralinguistics Challenge: Deception, Sincerity & Native Language (2016, background): “The con-ﬁguration ﬁle is the IS13 ComParE.conf, which is included in the 2.1 public release of openSMILE [28, 29].”
Improving Emotional TTS with an Emotion Intensity Input from Unsupervised Extraction (2021, methodology): “The openSMILE toolkit [14] is used to extract frame-level features (25 ms sliding window, 10ms shift).”
Multimodal Speech Emotion Recognition Using Audio and Text (2018, methodology): “All of these MFCC and prosodic features are extracted from the data using the OpenSMILE toolkit [27].”
Sentiment Adaptive End-to-End Dialog Systems (2018, methodology): “We used openSMILE (Eyben et al., 2013) to extract acoustic features.”
Word2VisualVec: Image and Video to Sentence Matching by Visual Feature Prediction (2016, methodology): “To exploit this channel, we extract a bag of quantized Mel-frequency Cepstral Coefficients (MFCC) vector [5] and concatenate it with the previous visual feature.”
Personalized machine learning for robot perception of affect and engagement in autism therapy (2018, methodology): “Robot sensing of outward and inward behavioral cues of the child used the open-source data processing tools: OpenFace (29), OpenPose (30), and openSMILE (31), together with tools that we built for processing audio-video and biosignals of the children (see note S3).”
AVEC 2017: Real-life Depression, and Affect Recognition Workshop and Challenge (2017, methodology): “e extraction of the LLDs and the functionals was done using the openSMILE4 toolkit [8]; for the BoAW generation, the openXBOW5 toolkit [24] was employed.”