Journal of New Music Research ( IF 1.1 ) Pub Date : 2021-01-19 , DOI: 10.1080/09298215.2021.1873393 C. Wick 1 , F. Puppe 1
The automatic recognition of scanned Medieval manuscripts written in square notation still represents a challenge due to degradation, non-standard layouts, or notations. We propose to apply CNN/LSTM networks that are trained using the segmentation-free CTC-loss-function. For evaluation, we use three different manuscripts and achieve a diplomatic Symbol Accuracy Rate (dSAR) of 86.0% on the most difficult book and 92.2% on the cleanest one. A neume dictionary during decoding yields a relative improvement of about 5%. Finally, we perform a thorough error analysis to provide a deeper insight into problems of the algorithm.
中文翻译:
使用CNN / LSTM网络和neume词典对中世纪音乐手稿的自动平方记号转录进行实验和详细的误差分析
由于降级,非标准版式或符号,自动识别以方形符号书写的中世纪扫描手稿仍然构成挑战。我们建议应用使用无分段CTC损失函数训练的CNN / LSTM网络。为了进行评估,我们使用三种不同的手稿,在最困难的书上达到了86.0%的外交符号准确率(dSAR),在最干净的书上达到了92.2%的外交准确率。解码期间的neume字典产生约5%的相对改善。最后,我们进行彻底的错误分析,以更深入地了解算法的问题。