The Speed Submission To DIHARD II: Contributions & Lessons Learned

Abstract

This paper describes the speaker diarization systems developed for the Second DIHARD Speech Diarization Challenge (DIHARD II) by the Speed team. Besides describing the system, which considerably outperformed the challenge baselines, we also focus on the lessons learned from numerous approaches that we tried for single and multi-channel systems. We present several components of our diarization system, including categorization of domains, speech enhancement, speech activity detection, speaker embeddings, clustering methods, resegmentation, and system fusion. We analyze and discuss the effect of each such component on the overall diarization performance within the realistic settings of the challenge.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keysahidullah2019the

Related papers

• DIHARD II Is Still Hard: Experimental Results And Discussions From The DKU-LENOVO Team (2020)6.34
• The Second DIHARD Diarization Challenge: Dataset, Task, And Baselines (2019)15.00
• UWB-NTIS Speaker Diarization System For The DIHARD II 2019 Challenge (2019)4.52
• The Dku-duke-lenovo System Description For The Third DIHARD Speech Diarization Challenge (2021)0.00
• The Hitachi-jhu DIHARD III System: Competitive End-to-end Neural Diarization And X-vector Clustering Systems Combined By Dover-lap (2021)0.00
• Enhancements For Audio-only Diarization Systems (2019)0.00
• BUT System Description For DIHARD Speech Diarization Challenge 2019 (2019)0.00
• The Royalflush Automatic Speech Diarization And Recognition System For In-car Multi-channel Automatic Speech Recognition Challenge (2024)0.00