Low-complexity Real-time Neural Network for Blind Bandwidth Extension of Wideband Speech

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference article in proceedingsScientificvertaisarvioitu

204 Lataukset (Pure)

Abstrakti

Speech is streamed at 16 kHz or lower sample rates in many applications (e.g. VoIP, Bluetooth headsets). Extending its bandwidth can produce significant quality improvements. We introduce BBWEXNet, a lightweight neural network that performs blind bandwidth extension of speech from 16 kHz (wideband) to 48 kHz (fullband) in real-time in CPU. Our low latency approach allows running the model with a maximum algorithmic delay of 16 ms, enabling end-to-end communication in streaming services and scenarios where the GPU is busy or unavailable. We propose a series of optimizations that take advantage of the U-Net architecture and vector quantization methods commonly used in speech coding, to produce a model whose performance is comparable to previous real-time solutions, but approximately halving the memory footprint and computational cost. Moreover, we show that the model complexity can be further reduced with a marginal impact on the perceived output quality.
AlkuperäiskieliEnglanti
Otsikko31st European Signal Processing Conference, EUSIPCO 2023 - Proceedings
KustantajaEuropean Association For Signal and Imag Processing
Sivut31-35
Sivumäärä5
ISBN (elektroninen)978-94-645936-0-0
DOI - pysyväislinkit
TilaJulkaistu - 4 syysk. 2023
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaEuropean Signal Processing Conference - Helsinki, Suomi
Kesto: 4 syysk. 20238 syysk. 2023
Konferenssinumero: 31
https://eusipco2023.org/

Julkaisusarja

NimiEuropean Signal Processing Conference
ISSN (elektroninen)2076-1465

Conference

ConferenceEuropean Signal Processing Conference
LyhennettäEUSIPCO
Maa/AlueSuomi
KaupunkiHelsinki
Ajanjakso04/09/202308/09/2023
www-osoite

Sormenjälki

Sukella tutkimusaiheisiin 'Low-complexity Real-time Neural Network for Blind Bandwidth Extension of Wideband Speech'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä