We present an end-to-end multichannel speaker-attributed automatic speech
recognition (MC-SA-ASR) system that combines a Conformer-based encoder with
multi-frame crosschannel attention and a speaker-attributed Transformer-based
decoder. To the best of our knowledge, this is the first model that efficiently
integrates ASR and speaker identification modules in a multichannel setting. On
simulated mixtures of LibriSpeech data, our system reduces the word error rate
(WER) by up to 12% and 16% relative compared to previously proposed
single-channel and multichannel approaches, respectively. Furthermore, we
investigate the impact of different input features, including multichannel
magnitude and phase information, on the ASR performance. Finally, our
experiments on the AMI corpus confirm the effectiveness of our system for
real-world multichannel meeting transcription.

我们提出了一种端到端的多通道说话人归属自动语音识别系统（MC-SA-ASR），它将基于 Conformer 的编码器与多帧跨通道注意力和基于说话人归属的 Transformer 解码器相结合。据我们所知，这是第一个在多通道环境中高效集成 ASR 和说话人识别模块的模型。在 LibriSpeech 数据的模拟混合语音中，与之前提出的单通道和多通道方法相比，我们的系统将词错误率（WER）相对降低了 12% 和 16%。此外，我们还研究了不同输入特征（包括多通道幅度和相位信息）对 ASR 性能的影响。最后，我们在 AMI 语料库上对我们的系统进行了实验，确认了我们系统在真实多通道会议转录中的有效性。