ASR-Based, Single-Ended Modeling of Listening Effort ? A Tool for TV Sound Engineers - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

ASR-Based, Single-Ended Modeling of Listening Effort ? A Tool for TV Sound Engineers

Hannah Baumgartner
  • Fonction : Auteur
Stefan Goetze
  • Fonction : Auteur
Jan Rennies
  • Fonction : Auteur

Résumé

The perceived speech intelligibility of many TV productions is a subject of complaints by TV viewers. In the production process of broadcast material, sound engineers mixing speech with background sounds (e.g. music or ?atmosphere?) typically listen to the sound material many times under ideal acoustical conditions with high-quality equipment, which is different from the listening situation of TV viewers. Hence, the sound engineer might underestimate the listener?s effort required to understand the speech. This contribution reviews our research approaches towards better understanding and eventually counteracting these problems, including series of psychoacoustic studies with normal-hearing and hearing-impaired listeners as well as different model approaches. The outcome of this research is a tool to automatically measure and display the perceived listening effort required to understand speech in a variety of different background sounds. It is single-ended, i.e. it does not require a clean speech reference, and is based on an automatic speech recognition (ASR) system. Speech distortions and interfering background sounds increase the uncertainty of the ASR system, which can be quantified and mapped to a perceptually interpretable scale using a psychoacoustic modeling approach. This performance measure correlates well with mean subjective listening effort ratings for a variety of distortions and acoustic backgrounds typical for TV broadcast material (r > 0.9). In principle, the tool is applicable to be integrated as a software plugin for digital audio workstations (DAWs) to support the work support of sound engineers, or in other applications such as speech quality monitoring of communication channels or real-time control of signal-enhancement algorithms.
Fichier principal
Vignette du fichier
000317.pdf (954.71 Ko) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte

Dates et versions

hal-03234203 , version 1 (26-05-2021)

Identifiants

Citer

Rainer Huber, Hannah Baumgartner, Stefan Goetze, Jan Rennies. ASR-Based, Single-Ended Modeling of Listening Effort ? A Tool for TV Sound Engineers. Forum Acusticum, Dec 2020, Lyon, France. pp.2441-2445, ⟨10.48465/fa.2020.0317⟩. ⟨hal-03234203⟩

Collections

FA2020
105 Consultations
84 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More