The importance of fillers for text representations of speech transcripts

09/23/2020

∙

While being an essential component of spoken language, fillers (e.g."um" or "uh") often remain overlooked in Spoken Language Understanding (SLU) tasks. We explore the possibility of representing them with deep contextualised embeddings, showing improvements on modelling spoken language and two downstream tasks - predicting a speaker's stance and expressed confidence.

READ FULL TEXT

The importance of fillers for text representations of speech transcripts

Improving Spoken Language Understanding By Exploiting ASR N-best Hypotheses

Weak-Supervised Dysarthria-invariant Features for Spoken Language Understanding using an FHVAE and Adversarial Training

Coupled Representation Learning for Domains, Intents and Slots in Spoken Language Understanding

You were saying? – Spoken Language in the V3C Dataset

Spoken Language Understanding on the Edge

Design Considerations For Hypothesis Rejection Modules In Spoken Language Understanding Systems

Finstreder: Simple and fast Spoken Language Understanding with Finite State Transducers using modern Speech-to-Text models

The importance of fillers for text representations of speech transcripts

Related Research

Improving Spoken Language Understanding By Exploiting ASR N-best Hypotheses

Weak-Supervised Dysarthria-invariant Features for Spoken Language Understanding using an FHVAE and Adversarial Training

Coupled Representation Learning for Domains, Intents and Slots in Spoken Language Understanding

You were saying? – Spoken Language in the V3C Dataset

Spoken Language Understanding on the Edge

Design Considerations For Hypothesis Rejection Modules In Spoken Language Understanding Systems

Finstreder: Simple and fast Spoken Language Understanding with Finite State Transducers using modern Speech-to-Text models