A new research paper explores input representations for multi-pitch estimation in vocal ensembles, a task complicated by overlapping vocal ranges and fundamental frequencies. The study compares the harmonic constant-Q transform (HCQT) with a linear short-time Fourier transform (STFT), finding that STFT outperforms HCQT while being less computationally expensive. The research suggests that finer frequency resolution is not always beneficial for this task and that shorter analysis windows can be more effective for time-varying vocal pitches. AI
IMPACT This research could lead to more efficient and accurate audio processing tools for music analysis and generation.
RANK_REASON Academic paper published on arXiv detailing a novel approach to a specific signal processing problem. [lever_c_demoted from research: ic=1 ai=0.7]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
