Maqam, a singing type, is a significant component of Kurdish music. A maqam singer receives training in a traditional face-to-face or through self-training. Automatic Singing Assessment (ASA) uses machine learning (ML) to provide the accuracy of singing styles and can help learners to improve their performance through error detection. Currently, the available ASA tools follow Western music rules. The musical composition requires all notes to stay within their expected pitch range from start to finish. The system fails to detect micro-intervals and pitch bends, so it identifies Kurdish maqam singing as incorrect even though the singer performs according to traditional rules. Kurdish maqam requires recognizing performance errors within microtonal spaces, which is beyond Western equal temperament. This research is the first attempt to address the mentioned gap. While many error types happen during singing, our focus is on pitch, rhythm, and modal stability errors in the context of Bayati-Kurd. We collected 50 songs from 13 vocalists ( 2-3 hours) and annotated 221 error spans (150 fine pitch, 46 rhythm, 25 modal drift). The data was segmented into 15,199 overlapping windows and converted to log-mel spectrograms. We developed a two-headed CNN-BiLSTM with attention mode to decide whether a window contains an error and to classify it based on the chosen errors. Trained for 20 epochs with early stopping at epoch 10, the model reached a validation macro-F1 of 0.468. On the full 50-song evaluation at a 0.750 threshold, recall was 39.4% and precision 25.8% . Within detected windows, type macro-F1 was 0.387, with F1 of 0.492 (fine pitch), 0.536 (rhythm), and 0.133 (modal drift); modal drift recall was 8.0%. The better performance on common error types shows that the method works, while the poor modal-drift recall shows that more data and balancing are needed.
Figures & tables
Figure 1: Overview of the research design we decided to use in this study.
Figure 2: The three error types: (a) fine pitch; (b) rhythm; (c) modal drift.
Figure 3: Model architecture with CNN feature extractor, bidirectional LSTM encoder, attention pooling, and two-task output heads.
Component
Details
Output size
Input
Log-mel spectrogram
1×128×T
CNN Blocks 1–4
3 × 3 Conv, BN, ReLU, MaxPool 2 × 2
128×8×T/16
BiLSTM
2 layers, 256 per direction, dropout 0.3
Sequence of 512 dims
Attention
512 → 256 (Tanh), dropout 0.2, 256 → 1, softmax
512 dims
Detection head
Dropout 0.3, Linear 512 → 1, sigmoid
1 (probability)
Type head
Linear 512 → 256, ReLU, dropout 0.3, Linear 256 → 3, softmax
3 classes
Table 1: Summary of the model architecture.
Figure 4: The Vocal Annotator interface used for creating training annotations.
Statistic
Value
Total songs
50
Total singers
13
Total annotations
221
Average errors per song
4.4
Songs with modal drift
8
Table 2: Dataset collection statistics.
Error type
Count
Percentage
Fine pitch error
150
67.9%
Rhythm error
46
20.8%
Modal drift error
25
11.3%
Total
221
100%
Table 3: Error type distribution in the collected dataset.
Figure 6: Error type distribution in the collected dataset (221 annotations).
Split
Windows
Percentage
Training
13,363
88.0%
Validation
718
4.7%
Test
1,118
7.4%
Total
15,199
100%
Table 4: Dataset split statistics.
Error type
Count
Percentage
Fine pitch error
537
67.5%
Rhythm error
212
26.6%
Modal drift error
47
5.9%
Total errors
796
–
Clean windows
12,567
94.0%
Table 5: Error distribution in training set.
Metric
Value
Detection threshold
0.750
Detection recall
39.4%
Detection precision
25.8%
Detection F1
0.311
Type macro-F1
0.387
Table 6: Comprehensive evaluation results on all 50 songs.
Error type
Detected
Precision
Recall
F1 score
Fine pitch error
57
89.5%
34.0%
0.492
Rhythm error
25
76.0%
41.3%
0.536
Modal drift error
5
40.0%
8.0%
0.133
Table 7: Per-class classification performance on all 50 songs.
Figure 7: Validation macro-F1 and detection F1 over 20 epochs. Best validation macro-F1 (0.468) at epoch 10.
Figure 8: Confusion matrix for error-type classification (87 matched detections). Diagonal cells (shaded) are correct predictions.