# Towards Leveraging Sequential Structure in Animal Vocalizations

arXiv: https://arxiv.org/abs/2511.10190 · HTML: https://mlsanity.com/paper/2511.10190/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 27 references that are in the map.
- First posted on arXiv 13 Nov 2025.

## Builds on (most cited first, 13 of 13)

- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [vq-wav2vec: Self-Supervised Learning of Discrete Speech Representations](https://mlsanity.com/paper/1910.05453/) (2019)
- [Exploring Speech Recognition, Translation, and Understanding with Discrete Speech Units: A Comparative Study](https://mlsanity.com/paper/2309.15800/) (2023)
- [Can Self-Supervised Neural Representations Pre-Trained on Human Speech distinguish Animal Callers?](https://mlsanity.com/paper/2305.14035/) (2023)
- [Comparing Self-Supervised Learning Models Pre-Trained on Human Speech and Animal Vocalizations for Bioacoustics Processing](https://mlsanity.com/paper/2501.05987/) (2025)
- [Towards Dog Bark Decoding: Leveraging Human Speech Processing for Automated Bark Classification](https://mlsanity.com/paper/2404.18739/) (2024)
- [On the Utility of Speech and Audio Foundation Models for Marmoset Call Analysis](https://mlsanity.com/paper/2407.16417/) (2024)
- [Feature Representations for Automatic Meerkat Vocalization Classification](https://mlsanity.com/paper/2408.15296/) (2024)
- [On feature representations for marmoset vocal communication analysis](https://mlsanity.com/paper/2504.14981/) (2025)

## Built on by (most cited first, 1 of 1)

- [BeeVe: Unsupervised Acoustic State Discovery in Honey Bee Buzzing](https://mlsanity.com/paper/2605.07903/) (2026)

## Abstract (arXiv)

Animal vocalizations contain sequential structures that carry important communicative information, yet most computational bioacoustics studies average the extracted frame-level features across the temporal axis, discarding the order of the sub-units within a vocalization. This paper investigates whether discrete acoustic token sequences, derived through vector quantization and gumbel-softmax vector quantization of extracted self-supervised speech model representations can effectively capture and leverage temporal information. To that end, pairwise distance analysis of token sequences generated from HuBERT embeddings shows that they can discriminate call-types and callers across four bioacoustics datasets. Sequence classification experiments using k-Nearest Neighbour with Levenshtein distance show that the vector-quantized token sequences yield reasonable call-type and caller classification performances, and hold promise as alternative feature representations towards leveraging sequential information in animal vocalizations.