近日，奇富科技智能语音团队论文《Qifusion-Net：基于特征融合的流式/非流式端到端语音识别框架》(Qifusion-Net: Layer-adapted Stream/Non-stream Model for End-to-End Multi-Accent Speech Recognition)被全球语音与声学顶级会议INTERSPEECH 2024收录。这是继去年《Eden-TTS：一种...