speech enhancement

DataFunSummit

Feb 4, 2024 · Mobile Development

Advanced Mobile Audio Recording Techniques in Quanjian K‑Song: Low Latency, High Fidelity, and Intelligent Audio Processing

The article details how Quanjian K‑Song has built a comprehensive mobile‑focused audio recording system since 2014, covering low‑latency capture, high‑quality sampling, lyric and vocal‑accompaniment alignment, ear‑return, pitch shifting, vocal enhancement, 3A processing, and AI‑driven scoring to deliver a professional karaoke experience on smartphones.

AI scoringAudio ProcessingLow Latency

0 likes · 14 min read

Advanced Mobile Audio Recording Techniques in Quanjian K‑Song: Low Latency, High Fidelity, and Intelligent Audio Processing

Kuaishou Tech

Dec 28, 2023 · Artificial Intelligence

Kuaishou Audio Team Wins ICASSP 2024 SSI and PLC Challenges with Advanced Speech Enhancement and Packet Loss Concealment

The Kuaishou audio team secured first place in both the ICASSP 2024 Speech Signal Improvement and Audio Deep Packet Loss Concealment challenges by deploying a two‑stage GAN‑based speech enhancement system and a hybrid time‑frequency packet‑loss concealment model that dramatically improve real‑time communication quality.

Audio ProcessingGANICASSP 2024

0 likes · 8 min read

Kuaishou Audio Team Wins ICASSP 2024 SSI and PLC Challenges with Advanced Speech Enhancement and Packet Loss Concealment

Kuaishou Tech

Feb 7, 2022 · Artificial Intelligence

Kuaishou Audio Team Wins ICASSP 2022 AEC Challenge Runner‑up and Publishes Two Deep Noise Suppression Papers

The Kuaishou audio team achieved a world‑runner‑up in the ICASSP 2022 Acoustic Echo Cancellation Challenge and had two deep noise suppression papers accepted, showcasing a hybrid traditional‑signal‑processing and deep‑learning system that excels in echo removal, noise reduction, and speech recognition across multiple real‑time communication scenarios.

Acoustic Echo CancellationArtificial IntelligenceDeep Noise Suppression

0 likes · 7 min read

Kuaishou Audio Team Wins ICASSP 2022 AEC Challenge Runner‑up and Publishes Two Deep Noise Suppression Papers

Tencent Cloud Developer

Mar 19, 2020 · Artificial Intelligence

Real-Time Voice Communication Technologies and AI Enhancements in Tencent Meeting

Shang Shidong outlines Tencent Meeting’s shift from analog PSTN to IP‑based VoIP, using H.323, SIP, RTP/UDP and the Opus codec, while AI‑driven super‑resolution, deep‑learning packet‑loss concealment, advanced noise reduction, and speech‑music classification boost audio quality, complemented by reference‑free MOS assessment and future 5G‑enabled cloud, IoT and WebRTC integration.

AIAudio ProcessingRTP

0 likes · 30 min read

Real-Time Voice Communication Technologies and AI Enhancements in Tencent Meeting