On the Use of Audio to Improve Dialogue Policies

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Roncel, Daniel, Costa, Federico, Hernando, Javier
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929547850547200
author Roncel, Daniel
Costa, Federico
Hernando, Javier
author_facet Roncel, Daniel
Costa, Federico
Hernando, Javier
contents With the significant progress of speech technologies, spoken goal-oriented dialogue systems are becoming increasingly popular. One of the main modules of a dialogue system is typically the dialogue policy, which is responsible for determining system actions. This component usually relies only on audio transcriptions, being strongly dependent on their quality and ignoring very important extralinguistic information embedded in the user's speech. In this paper, we propose new architectures to add audio information by combining speech and text embeddings using a Double Multi-Head Attention component. Our experiments show that audio embedding-aware dialogue policies outperform text-based ones, particularly in noisy transcription scenarios, and that how text and audio embeddings are combined is crucial to improve performance. We obtained a 9.8% relative improvement in the User Request Score compared to an only-text-based dialogue system on the DSTC2 dataset.
format Preprint
id arxiv_https___arxiv_org_abs_2410_13385
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle On the Use of Audio to Improve Dialogue Policies
Roncel, Daniel
Costa, Federico
Hernando, Javier
Audio and Speech Processing
Computation and Language
Sound
With the significant progress of speech technologies, spoken goal-oriented dialogue systems are becoming increasingly popular. One of the main modules of a dialogue system is typically the dialogue policy, which is responsible for determining system actions. This component usually relies only on audio transcriptions, being strongly dependent on their quality and ignoring very important extralinguistic information embedded in the user's speech. In this paper, we propose new architectures to add audio information by combining speech and text embeddings using a Double Multi-Head Attention component. Our experiments show that audio embedding-aware dialogue policies outperform text-based ones, particularly in noisy transcription scenarios, and that how text and audio embeddings are combined is crucial to improve performance. We obtained a 9.8% relative improvement in the User Request Score compared to an only-text-based dialogue system on the DSTC2 dataset.
title On the Use of Audio to Improve Dialogue Policies
topic Audio and Speech Processing
Computation and Language
Sound
url https://arxiv.org/abs/2410.13385