Till KTH:s startsida Till KTH:s startsida

Seminar: Front-end audio processing for hands-free speech communication

Tid: Måndag 22 februari 2016 kl 17:00 - 19:00 2016-02-22T17:00:00 2016-02-22T19:00:00

Kungliga Tekniska högskolan
VT 2016

Plats: Q2

Aktivitet: Seminarium

Studentgrupper: TIKTM_1, TIKTM_2, TIVNM_DMTE_1, TIVNM_DMTE_2, TIVNM_ITAK_2, TTLSM_1, TTLSM_2

Info:

Speaker: Christian Schuldt, Limes Audio

Abstract:

In hands-free speech communication front-ends such as, e.g., conference phones, the microphone generally picks a desired speech signal that has been degraded by reverberation caused by the room, together with disturbing background noise and significant echo from the loudspeaker. Hence, there are three major problems that needs to be tackled: reverberation, noise and echo. This presentation will address ways of dealing with these problems, and the focus will be on practical signal processing aspects.

Schemahandläggare skapade händelsen 23 september 2015

ändrade rättigheterna 3 november 2015

Kan därmed läsas av alla och ändras av lärare.
Schemahandläggare tog bort händelsen 13 november 2015
Lärare Lars Kildehøj redigerade 13 januari 2016

Seminarium: Front-end audio processing for hands-free speech communication

Speaker: Christian Schuldt, Limes Audio¶

Abstract: ¶

In hands-free speech communication front-ends such as, e.g., conference phones, the microphone generally picks a desired speech signal that has been degraded by reverberation caused by the room, together with disturbing background noise and significant echo from the loudspeaker. Hence, there are three major problems that needs to be tackled: reverberation, noise and echo. This presentation will address ways of dealing with these problems, and the focus will be on practical signal processing aspects.¶

Hela världen får läsa.

Senast ändrad 2016-01-13 14:02

Taggar: Saknas än så länge.