Loudspeaker and Listening Position Estimation Using Smart Speakers

Abstract

Recently, so-called smart speakers have been introduced and they include a microphone array. One potential application of such a smart speaker is to use it for calibrating a larger audio system which the speaker is a part of. In this paper, we propose a method to perform this calibration using one or several smart speakers. Specifically, a map is estimated of the sensors and sound sources. As opposed to existing methods, the proposed method can create this map for both synchronised and unsynchronised sound sources by taking the different localisation errors into account. We show that this gives more accurate estimates than assuming identical estimation errors, and that existing methods are outperformed in terms of estimation accuracy for various noise levels and reverberation times.

Speaker Distance Estimation in Enclosures From Single-Channel Audio

“…Extensive research has been conducted on this subject for various acoustic systems that commonly use distributed microphone arrays. These systems encompass a range of setups, such as intelligent loudspeakers [12], spherical microphones [13], triangular configu-rations [14], and arrays of acoustic sensors [15]. Simpler audio formats including binaural recordings have been investigated to a much lesser extent, including few studies with classical machine learning methods [4], [16] and very limited research related to deep learning [7], [8].…”

Binaural Sound Source Distance Estimation and Localization for a Moving Listener

“…Joint SDE and DOAE or SDEL is usually defined as position estimation, referring to either a continuous position coordinate in 2D/3D space (e.g., expressed in Cartesian x, y, z coordinates) or to pre-defined position classes corresponding to a spatial "binning" of the region of interest. This topic has been widely researched for multiple types of acoustic systems employing typically distributed microphone arrays, including spherical microphones [50], triangular [51], smart loudspeakers [52] or acoustic sensor arrays [53]. However, only a few studies aimed at position estimation from binaural recordings.…”

Affordances and Constraints of Intelligent Personal Assistants for Second-Language Learning

“…This can consist of basic tasks – that is, transactional interactions involving searching for information, setting reminders, and playing music via a smartphone or smart speaker. However, in a survey focused on user behavior with smart speakers, Nielsen (2018) found that 68% of users chat with an IPA for fun, which indicates that they are being used for more than just transactional interactions – that is, social interactions with virtual assistants are becoming more commonplace. Relatedly, the chat bots developed through the Alexa Prize, a competition to promote the development of conversational artificial intelligence, and Google’s Meena are examples of social bots that were developed with the aim of enabling users to have engaging, open-ended conversations with a virtual assistant.…”