Towards Efficient And Real-time Piano Transcription Using Neural Autoregressive Models

Abstract

In recent years, advancements in neural network designs and the availability of large-scale labeled datasets have led to significant improvements in the accuracy of piano transcription models. However, most previous work focused on high-performance offline transcription, neglecting deliberate consideration of model size. The goal of this work is to implement real-time inference for piano transcription while ensuring both high performance and lightweight. To this end, we propose novel architectures for convolutional recurrent neural networks, redesigning an existing autoregressive piano transcription model. First, we extend the acoustic module by adding a frequency-conditioned FiLM layer to the CNN module to adapt the convolutional filters on the frequency axis. Second, we improve note-state sequence modeling by using a pitchwise LSTM that focuses on note-state transitions within a note. In addition, we augment the autoregressive connection with an enhanced recursive context. Using thes

Stats

• citations5
• S2 citations—
• github stars0
• HF likes0
• heat score5.84
• arxiv keykwon2024towards

Related papers

• Hppnet: Modeling The Harmonic Structure And Pitch Invariance In Piano Transcription (2022)0.00
• Deep Polyphonic ADSR Piano Note Transcription (2019)11.29
• Audio-to-score Alignment Of Piano Music Using Rnn-based Automatic Music Transcription (2017)0.00
• Piano Transcription By Hierarchical Language Modeling With Pretrained Roll-based Encoders (2025)4.52
• Invariances And Data Augmentation For Supervised Music Transcription (2017)11.08
• Streaming Piano Transcription Based On Consistent Onset And Offset Decoding With Sustain Pedal Detection (2025)0.00
• D3RM: A Discrete Denoising Diffusion Refinement Model For Piano Transcription (2025)5.93
• Adversarial Learning For Improved Onsets And Frames Music Transcription (2019)0.00