The PCG-AIID System for L3DAS22 Challenge: MIMO and MISO convolutional recurrent Network for Multi Channel Speech Enhancement and Speech Recognition
This paper described the PCG-AIID system for L3DAS22 challenge in Task 1: 3D speech enhancement in office reverberant environment. We proposed a two-stage framework to address multi-channel speech denoising and dereverberation. In the first stage, a multiple input and multiple output (MIMO) network is applied to remove background noise while maintaining the spatial characteristics of multi-channel signals. In the second stage, a multiple input and single output (MISO) network is applied to enhance the speech from desired direction and post-filtering. As a result, our system ranked 3rd place in ICASSP2022 L3DAS22 challenge and significantly outperforms the baseline system, while achieving 3.2% WER and 0.972 STOI on the blind test-set.
Code (0)
등록된 구현이 없습니다.
Tasks
DenoisingSpeech DenoisingSpeech Enhancementspeech-recognitionSpeech RecognitionSimilar Papers 제목 키워드 기반
MIMO-DoAnet: Multi-channel Input and Multiple Outputs DoA Network with Unknown Number of Sound Sources
Recent neural network based Direction of Arrival (DoA) estimation algorithms have performed well on unknown number of sound sources scenarios. These algorithms are usually achieved by mapping the multi-channel audio inpu…
Range Resolution of Near-field MIMO Sensing
The radiative near-field and integration of sensing capabilities are seen as two key components of the next generation of wireless communication systems. In this paper, the sensing performance of a narrowband near-field …
MIMONet: Multi-Input Multi-Output On-Device Deep Learning
Future intelligent robots are expected to process multiple inputs simultaneously (such as image and audio data) and generate multiple outputs accordingly (such as gender and emotion), similar to humans. Recent research h…
Deep LearningModel CompressionRIS-aided Joint Localization and Synchronization with a Single-Antenna MmWave Receiver
MmWave multiple-input single-output (MISO) systems using a single-antenna receiver are regarded as promising solution for the near future, before the full-fledged 5G MIMO will be widespread. However, for MISO systems syn…
PositionIndexing AI Risks with Incidents, Issues, and Variants
Two years after publicly launching the AI Incident Database (AIID) as a collection of harms or near harms produced by AI in the world, a backlog of "issues" that do not meet its incident ingestion criteria have accumulat…
Computer Security