Phone Segmentation and Recognition through Phonological Activation Mapping

Hugging Face Daily Papers Papers
phone-segmentation speech-recognition self-supervised-learning phonological-activation lightweight-model

Summary

This paper introduces SPAM (S3M-based Phonological Activation Mapping), a method that leverages self-supervised speech models to perform both phone segmentation and recognition simultaneously using lightweight, gradient-descent-free prediction heads requiring minimal phonetic transcriptions.

Phone segmentation and recognition are inherently related tasks, yet modern approaches typically model them separately. We argue that phonetic structure is already latent in the representations of self-supervised speech models (S3Ms), and one only needs to steer them to solve both tasks. We leverage S3M-based Phonological Activation Mapping (SPAM), which maps each S3M representation frame to a vector of phonological feature activations, such as voicing and nasality. On top of SPAM, we introduce two simple but effective lightweight, gradient-descent-free prediction heads: a recognition head and a segmentation head. Our method requires less than a minute of phonetic transcriptions, and generalizes to unseen phones during training. Across a diverse range of datasets, our approach attains strong segmentation and recognition performance.
Original Article
View Cached Full Text

Cached at: 07/13/26, 07:49 AM

Paper page - Phone Segmentation and Recognition through Phonological Activation Mapping

Source: https://huggingface.co/papers/2607.09020 Authors:

,

,

,

,

,

,

,

,

,

Abstract

Phonesegmentationandrecognitionareinherentlyrelatedtasks,yetmodernapproachestypicallymodelthemseparately.Wearguethatphoneticstructureisalreadylatentintherepresentationsofself-supervisedspeechmodels(S3Ms),andoneonlyneedstosteerthemtosolvebothtasks.WeleverageS3M-basedPhonologicalActivationMapping(SPAM),whichmapseachS3Mrepresentationframetoavectorofphonologicalfeatureactivations,suchasvoicingandnasality.OntopofSPAM,weintroducetwosimplebuteffectivelightweight,gradient-descent-freepredictionheads:arecognitionheadandasegmentationhead.Ourmethodrequireslessthanaminuteofphonetictranscriptions,andgeneralizestounseenphonesduringtraining.Acrossadiverserangeofdatasets,ourapproachattainsstrongsegmentationandrecognitionperformance.

View arXiv pageView PDFGitHub1Add to collection

Get this paper in your agent:

hf papers read 2607\.09020

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2607.09020 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2607.09020 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.09020 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

SAM 3: Segment Anything with Concepts

Papers with Code Trending

SAM 3 introduces a unified model for promptable concept segmentation and tracking, achieving state-of-the-art performance with a decoupled recognition and localization architecture and a scalable data engine.

Phonological Perception of Sign Language Models

arXiv cs.CL

This paper evaluates whether Sign Language Recognition models exhibit phonological sensitivity by probing them with minimal pairs of signs, revealing architectural trade-offs and emergent but limited phonological perception.