Guitar chord recognition based on finger patterns with deep learning

Ooaku, Takumi; Linh, Tran Duy; Arai, Masayuki; Maekawa, Tsukasa; Mizutani, Kozo

doi:10.1145/3290420.3290422

Cited by 5 publications

(6 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using a slightly different methodology for chord classification Takumi, Tran [29] proposed a method using neural networks and machine learning to recognize chords played despite background noise. The attempt is to build a system that can realize the finger patterns of guitar players in video and can automatically generate the corresponding chord played.…”

Section: Neural Network and Deep Learningmentioning

confidence: 99%

See 1 more Smart Citation

Note Frequency Recognition and Finger Motion Capture of a Guitarist: A Survey and Challenges Ahead

Emad

Sobeah

Hassan

et al. 2022

Journal of Computing and Communication

View full text Add to dashboard Cite

One of the main issues that face new guitar aspirants is that there is a lot of further information, which proves a lot to take in for just a beginner as it causes much-disliked confusion. Students also face problems with their left-hand motion and correct pitch frequency. Researchers have tackled this problem in many ways and have mainly landed on using two approaches. The first was Finger Motion Capture, where previous research specialized in analyzing images and videos of the guitarist. The second was Note Frequency Recognition, where the research's primary purpose was to examine the sound and audio recorded by the guitarist. This paper surveys all the approaches taken to solve this problem, discussing them in detail and exploring the challenges faced with each approach. Furthermore, this paper proposes a hybrid solution that includes both Note Frequency Recognition and Finger Motion Capture to make a full assessment and give feedback on a guitarist's performance.

show abstract

Section: Neural Network and Deep Learningmentioning

confidence: 99%

“…The main challenge faced by researchers when using this approach was that despite how advanced this approach is compared to others, working with multiple varying data may cause a significant decrease in the recognition accuracy [29].…”

Section: Neural Network and Deep Learningmentioning

confidence: 99%

Note Frequency Recognition and Finger Motion Capture of a Guitarist: A Survey and Challenges Ahead

Emad

Sobeah

Hassan

et al. 2022

Journal of Computing and Communication

View full text Add to dashboard Cite

show abstract

“…e method employs the pitch class profile (PCP), a well-known function vector for automatic chord recognition. While we are going to try a different approach with our model, the data used for the training of the model in this research paper is an invaluable addition to the dataset [17][18][19]. e dataset used in this paper consists of guitar chords recorded both in a studio (anechoic chamber) and in noisy environments.…”

Section: Neural Network For Musical Chordmentioning

confidence: 99%

Transfer Learning for Audio Waveform to Guitar Chord Spectrograms Using the Convolution Neural Network

Jadhav

Patel²,

Jhaveri

et al. 2022

Mobile Information Systems

View full text Add to dashboard Cite

Automatic chord recognition has always been approached as a broad music audition task. The desired output is a succession of time-aligned discrete chord symbols, such as GMaj and Asus2. Automatic music transcription is the process of converting a musical recording into a human-readable and interpretable representation. When dealing with polyphonic sounds or removing certain limits, automatic music transcription remains a difficult undertaking. A guitar, for example, presents a greater challenge, as guitarists can play the same note in a variety of places. The study makes use of CNN functionality to generate the guitar tab; initially, the constant-Q transform was used to turn the input audio file into short time spectrograms that the CNN model utilises to analyse the chord. The paper developed a method for extracting chord sequences and notes from audio recordings of solo guitar performances. For intervals in the supplied audio, the proposed approach outputs chord names and fret-board notes. The model described here has been refined to achieve an accuracy of 88.7%. The model’s ability to properly tag audio clips is an incredible advancement.

show abstract

“…In addition, in this sense, the authors have tried to avoid the traditional side of pure education [24]. The authors of this paper have opted to use a model that focuses on the user and the training process [25], rather than using models based on deep learning that require complicated cognitive processes.…”

Section: Objectives and Hypothesismentioning

confidence: 99%

“…Sci. 2020, 10, x FOR PEER R EVIEW 4 of 15 and the training process [25], rather than using models based on deep learning that require complicated cognitive processes.…”

Section: Conceptual Designmentioning

confidence: 99%

Augmented Reality to Facilitate Learning of the Acoustic Guitar

et al. 2020

View full text Add to dashboard Cite

Many people wishing to learn a musical instrument opt to learn using alternative or informal methods instead of the traditional Master–Apprentice model that requires a greater cognitive load. This paper presents an augmented reality (AR)-based application designed to teach and train guitar chords, with the novelty that it is also used to teach short melodies consisting of four chord transitions so that users have to change hand and finger positions. The app uses high-quality 3D models of an acoustic guitar and animated hand to indicate correct finger positions and the movements required when changing from one chord to another. To follow the animated instructions, the learner overlaps the 3D model onto the neck of the physical guitar and his or her own hand. A system usability scale (SUS) questionnaire was used to measure the usability of the application. A score of 82.0 was obtained, which is higher than the average of 68 points that indicates the application is good from a user experience perspective, thus satisfying the purpose for which it was created. Having analysed the data for both groups—individuals with no prior experience of playing a musical instrument versus individuals with prior experience—it was concluded that the application provided a useful learning approach for all participants involved in the study, regardless of experience. That said, those possessing prior experience of playing an instrument learnt faster. It should be noted that the research revealed significant difference in learning by gender, with male participants learning faster than female participants. Similar results have been detected in other research performed in the field of music, as well as in other fields. As this study required spatial reasoning when viewing the 3D model, the differences identified this case may well have arisen as a consequence of differences in men and women’s spatial awareness, thereby leaving open an alternative line of research.

show abstract

Guitar chord recognition based on finger patterns with deep learning

Cited by 5 publications

References 14 publications

Note Frequency Recognition and Finger Motion Capture of a Guitarist: A Survey and Challenges Ahead

Note Frequency Recognition and Finger Motion Capture of a Guitarist: A Survey and Challenges Ahead

Transfer Learning for Audio Waveform to Guitar Chord Spectrograms Using the Convolution Neural Network

Augmented Reality to Facilitate Learning of the Acoustic Guitar

Contact Info

Product

Resources

About