在當代訊號處理與人工智慧應用領域中，頻譜圖（Spectrogram）作為一種將時間域訊號轉換為時頻域的視覺化表示方式，已廣泛應用於人體動作辨識、生物醫學分析、語音識別以及環境聲音分類等多種研究領域。其中，梅爾頻譜圖（Mel-Spectrogram）作為頻譜圖的一種重要衍生，因其模擬人耳對頻率的感知特性，對頻率軸進行非線性壓縮，能更有效保留語音訊號中的語意與韻律結構，已成為目前最常用且具表達力的聲學特徵之一。透過直觀且細膩的時頻資訊呈現，頻譜圖可有效揭露訊號中潛在的時變頻率特性，並為深度學習模型提供高辨識度的輸入特徵，特別是在卷積神經網路（CNN）架構中展現出卓越的分類與辨識能力。

本論文基於上述兩種頻譜圖，提出了一種自適應頻率軸頻譜圖（Adaptive Frequency Spectrogram, Adaf-Spectrogram）。該頻譜圖透過計算整體資料集的頻率能量分布，自動調整頻率軸的尺度縮放，以更有效地突顯訊號中的關鍵頻率特徵。實驗結果證明，此自適應頻率軸頻譜圖在多種資料集上均具良好適應性，並且在辨識效果上優於傳統頻譜圖（Spectrogram），展現出顯著的性能提升。

In the fields of modern signal processing and artificial intelligence, the spectrogram is a fundamental visual representation that transforms time-domain signals into the time-frequency domain. It has found extensive applications in areas such as human activity recognition, biomedical signal analysis, speech recognition, and environmental sound classification. Among these, the Mel-spectrogram is a prominent variant. By emulating the human auditory system's perception of frequency through non-linear compression of the frequency axis, it more effectively preserves semantic and prosodic information in speech signals. Consequently, it has become one of the most expressive and widely-adopted acoustic features.
With its intuitive yet detailed time-frequency representation, the spectrogram effectively reveals latent time-variant frequency characteristics within a signal, providing highly discriminative input features for deep learning models. It has demonstrated exceptional performance in classification and recognition tasks, particularly within Convolutional Neural Network (CNN) architectures.

Building upon these established representations, this paper proposes a novel Adaptive Frequency Spectrogram (Adaf-Spectrogram). This data-driven method automatically adjusts the frequency axis scaling by computing the overall frequency energy distribution across an entire dataset, thereby more effectively emphasizing critical frequency features. Experimental results demonstrate that the proposed Adaf-Spectrogram exhibits excellent adaptability across multiple datasets. Furthermore, it outperforms conventional linear-scale spectrograms in recognition tasks, showcasing a significant performance improvement.

[1] M. Bellanger, “Digital processing of speech signals,” IEEE Transactions on Acous-
tics, Speech, and Signal Processing, vol. 28, no. 1, pp. 118–119, 1980. doi: 10.1109/
TASSP.1980.1163362.
[2] U. Saha, S. Saha, T. Kabir, S. A. Fattah, and M. Saquib, Decoding human activities:
Analyzing wearable accelerometer and gyroscope data for activity recognition, 2024.
arXiv: 2310.02011 [cs.CV]. [Online]. Available: https://arxiv.org/abs/2310.
02011.
[3] J. Vaghasiya, C. Mayorga-Martinez, and M. Pumera, “Wearable sensors for tele-
health based on emerging materials and nanoarchitectonics,” npj Flexible Electron-
ics, vol. 7, p. 26, 2023. doi: 10.1038/s41528-023-00261-4. [Online]. Available:
https://doi.org/10.1038/s41528-023-00261-4.
[4] A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep
convolutional neural networks,” in Advances in Neural Information Processing Sys-
tems, F. Pereira, C. Burges, L. Bottou, and K. Weinberger, Eds., vol. 25, Curran
Associates, Inc., 2012. [Online]. Available: https://proceedings.neurips.cc/
paper_files/paper/2012/file/c399862d3b9d6b76c8436e924a68c45b-Paper.
pdf.
[5] K. Simonyan and A. Zisserman, Very deep convolutional networks for large-scale
image recognition, 2015. arXiv: 1409.1556 [cs.CV]. [Online]. Available: https:
//arxiv.org/abs/1409.1556.
[6] A. Vaswani, N. Shazeer, N. Parmar, et al., “Attention is all you need,” in Ad-
vances in Neural Information Processing Systems, I. Guyon, U. V. Luxburg, S.
Bengio, et al., Eds., https://proceedings.neurips.cc/paper_files/paper/
2017 / file / 3f5ee243547dee91fbd053c1c4a845aa - Paper . pdf, vol. 30, Curran
Associates, Inc., 2017.
[7] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, Bert: Pre-training of deep
bidirectional transformers for language understanding, 2019. arXiv: 1810 . 04805
[cs.CL]. [Online]. Available: https://arxiv.org/abs/1810.04805.
[8] Y. LeCun and Y. Bengio, “Convolutional networks for images, speech, and time
series,” in The Handbook of Brain Theory and Neural Networks. Cambridge, MA,
USA: MIT Press, 1998, pp. 255–258, isbn: 0262511029.
[9] X. Shi, Z. Chen, H. Wang, D.-Y. Yeung, W.-k. Wong, and W.-c. Woo, Convolutional
lstm network: A machine learning approach for precipitation nowcasting, 2015.
arXiv: 1506.04214 [cs.CV]. [Online]. Available: https://arxiv.org/abs/1506.
04214.
[10] Z. Wang, W. Yan, and T. Oates, Time series classification from scratch with deep
neural networks: A strong baseline, 2016. arXiv: 1611.06455 [cs.LG]. [Online].
Available: https://arxiv.org/abs/1611.06455.
[11] H. Ismail Fawaz, G. Forestier, J. Weber, L. Idoumghar, and P.-A. Muller, “Deep
learning for time series classification: A review,” Data Mining and Knowledge Dis-
covery, vol. 33, no. 4, pp. 917–963, Mar. 2019, issn: 1573-756X. doi: 10.1007/
s10618-019-00619-1. [Online]. Available: http://dx.doi.org/10.1007/s10618-
019-00619-1.
[12] G. Zerveas, S. Jayaraman, D. Patel, A. Bhamidipaty, and C. Eickhoff, A transformer-
based framework for multivariate time series representation learning, 2020. arXiv:
2010.02803 [cs.LG]. [Online]. Available: https://arxiv.org/abs/2010.02803.
[13] L. Cohen, Time-Frequency Analysis. Upper Saddle River, NJ: Prentice Hall, 1995.
[14] L. Tang, Y. Jia, Y. Qian, S. Yi, and P. Yuan, “Human activity recognition based on
mixed cnn with radar multi-spectrogram,” IEEE Sensors Journal, vol. 21, no. 22,
pp. 25 950–25 962, 2021. doi: 10.1109/JSEN.2021.3118836.
[15] L. Qu, Y. Wang, T. Yang, and Y. Sun, “Human activity recognition based on wrgan-
gp-synthesized micro-doppler spectrograms,” IEEE Sensors Journal, vol. 22, no. 9,
pp. 8960–8973, 2022. doi: 10.1109/JSEN.2022.3164152.
[16] J. Huang, B. Chen, B. Yao, and W. He, “Ecg arrhythmia classification using
stft-based spectrogram and convolutional neural network,” IEEE Access, vol. 7,
pp. 92 871–92 880, 2019. doi: 10.1109/ACCESS.2019.2928017.
[17] S. Y. ŞEN and N. ÖZKURT, “Ecg arrhythmia classification by using convolutional
neural network and spectrogram,” in 2019 Innovations in Intelligent Systems and
Applications Conference (ASYU), 2019, pp. 1–6. doi: 10.1109/ASYU48272.2019.
8946417.
[18] H. Nyquist, “Certain topics in telegraph transmission theory,” Transactions of the
American Institute of Electrical Engineers, vol. 47, pp. 617–644, 1928.
[19] A. V. Oppenheim and R. W. Schafer, Discrete-Time Signal Processing, 2nd. Upper
Saddle River, NJ: Prentice Hall, 1999.
[20] S. S. Stevens, J. Volkmann, and E. B. Newman, “A scale for the measurement of
the psychological magnitude of pitch,” The Journal of the Acoustical Society of
America, vol. 8, no. 3, pp. 185–190, 1937.
[21] K. Choi, G. Fazekas, K. Cho, and M. Sandler, A tutorial on deep learning for
music information retrieval, 2018. arXiv: 1709.04396 [cs.CV]. [Online]. Available:
https://arxiv.org/abs/1709.04396.
[22] K. J. Piczak, “Environmental sound classification with convolutional neural net-
works,” in 2015 IEEE 25th International Workshop on Machine Learning for Signal
Processing (MLSP), 2015, pp. 1–6. doi: 10.1109/MLSP.2015.7324337.
[23] D. Issa, M. F. Demirci, and A. Yazici, “Speech emotion recognition with deep
convolutional neural networks,” Biomedical Signal Processing and Control, vol. 59,
p. 101 894, 2020, issn: 1746-8094. doi: 10.1016/j.bspc.2020.101894.
[24] L. Abdel Khaliq, S. Janzen, and W. Maass, “Reaver: Real-time earthquake pre-
diction with attention-based sliding-window spectrograms,” in Proceedings of the
Thirty-Third International Joint Conference on Artificial Intelligence, IJCAI-24,
K. Larson, Ed., Demo Track, International Joint Conferences on Artificial Intelli-
gence Organization, Aug. 2024, pp. 8596–8600. doi: 10.24963/ijcai.2024/988.
[Online]. Available: https://doi.org/10.24963/ijcai.2024/988.
[25] X. Li, G. A. Ng, and F. S. Schlindwein, “Transfer learning in heart sound classifi-
cation using mel spectrogram,” in 2022 Computing in Cardiology (CinC), vol. 498,
2022, pp. 1–4. doi: 10.22489/CinC.2022.046.
[26] M. Quadrini, E. Abdullah, N. Francioni, et al., “Sleep apnea detection using mel-
spectrograms snoring and convolutional neural networks,” in 2024 IEEE Interna-
tional Conference on Bioinformatics and Biomedicine (BIBM), 2024, pp. 6506–
6512. doi: 10.1109/BIBM62325.2024.10822147.
[27] W. Koenig, H. K. Dunn, and L. Y. Lacy, “The sound spectrograph,” Journal of the
Acoustical Society of America, vol. 18, no. 1, pp. 19–49, 1946.
[28] P. Arun, S. A. Lincon, and N. Prabhakaran, “An automated method for the analysis
of bearing vibration based on spectrogram pattern matching,” Journal of Applied
Research and Technology, vol. 17, no. 2, 2019. doi: 10.22201/icat.16656423.
2019 . 17 . 2 . 805. [Online]. Available: https : / / doi . org / 10 . 22201 / icat .
16656423.2019.17.2.805.
[29] N. A. Tawhid, S. Siuly, H. Wang, F. Whittaker, K. Wang, and Y. Zhang, “A spec-
trogram image based intelligent technique for automatic detection of autism spec-
trum disorder from eeg,” PLOS ONE, vol. 16, no. 6, e0253094, 2021, https://jour-
nals.plos.org/plosone/article/file?id=10.1371/journal.pone.0253094type=printable.
doi: 10 . 1371 / journal . pone . 0253094. [Online]. Available: https : / / app .
dimensions.ai/details/publication/pub.1139158677.
[30] D.-H. Park, M.-W. Jeon, and H.-N. Kim, Resolution-adaptive micro-doppler spectro-
gram for human activity recognition, 2024. arXiv: 2411.15057 [eess.SP]. [Online].
Available: https://arxiv.org/abs/2411.15057.
[31] S. Davis and P. Mermelstein, “Comparison of parametric representations for mono-
syllabic word recognition in continuously spoken sentences,” IEEE Transactions on
Acoustics, Speech, and Signal Processing, vol. 28, no. 4, pp. 357–366, 1980. doi:
10.1109/TASSP.1980.1163420.
[32] J. C. Brown, “Calculation of a constant q spectral transform,” The Journal of the
Acoustical Society of America, vol. 89, no. 1, pp. 425–434, 1991.
[33] A. Dosovitskiy, L. Beyer, A. Kolesnikov, et al., An image is worth 16x16 words:
Transformers for image recognition at scale, 2021. arXiv: 2010.11929 [cs.CV].
[Online]. Available: https://arxiv.org/abs/2010.11929.
[34] J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei, “Imagenet: A large-
scale hierarchical image database,” in 2009 IEEE Conference on Computer Vision
and Pattern Recognition, 2009, pp. 248–255. doi: 10.1109/CVPR.2009.5206848.
[35] K. He, X. Zhang, S. Ren, and J. Sun, Deep residual learning for image recognition,
2015. arXiv: 1512.03385 [cs.CV]. [Online]. Available: https://arxiv.org/abs/
1512.03385.
[36] M. Tan and Q. V. Le, Efficientnet: Rethinking model scaling for convolutional neural
networks, 2020. arXiv: 1905.11946 [cs.LG]. [Online]. Available: https://arxiv.
org/abs/1905.11946.
[37] L. Pham, P. Lam, T. Nguyen, H. Nguyen, and A. Schindler, “Deepfake audio de-
tection using spectrogram-based feature and ensemble of deep learning models,” in
2024 IEEE 5th International Symposium on the Internet of Sounds (IS2), 2024,
pp. 1–5. doi: 10.1109/IS262782.2024.10704095.
[38] Y. Gong, Y.-A. Chung, and J. Glass, Ast: Audio spectrogram transformer, 2021.
arXiv: 2104.01778 [cs.SD]. [Online]. Available: https://arxiv.org/abs/2104.
01778.
[39] Instituto Geofísico –EPN, Formulario de registros acelerográficos (señales sísmi-
cas), https://www.igepn.edu.ec/senales- sismicas/fomulario- eseismic,
Accessed: 2025-07-23, 2025.
[40] A. Salazar, R. Arroyo, N. Pérez, and D. Benítez, “Deep-learning for volcanic seis-
mic events classification,” in 2020 IEEE Colombian Conference on Applications of
Computational Intelligence (IEEE ColCACI 2020), 2020, pp. 1–6. doi: 10.1109/
ColCACI50549.2020.9247848.
[41] K. J. Piczak, “Esc: Dataset for environmental sound classification,” in Proceed-
ings of the 23rd ACM International Conference on Multimedia, ser. MM ’15,
Brisbane, Australia: Association for Computing Machinery, 2015, pp. 1015–1018,
isbn: 9781450334594. doi: 10.1145/2733373.2806390. [Online]. Available: https:
//doi.org/10.1145/2733373.2806390.
[42] V. Shanbhag, Bird song data set, https://www.kaggle.com/datasets/vinayshanbhag/
bird-song-data-set, Accessed: 2025-07-23, 2022.