SearcharxivSearch

arXiv · 2301.05295

Rock Guitar Tablature Generation via Natural Language Processing

Abstract

Deep learning has recently empowered and democratized generative modeling of images and text, with additional concurrent works exploring the possibility of generating more complex forms of data, such as audio. However, the high dimensionality, long-range dependencies, and lack of standardized datasets currently makes generative modeling of audio and music very challenging. We propose to model music as a series of discrete notes upon which we can use autoregressive natural language processing techniques for successful generative modeling. While previous works used similar pipelines on data such as sheet music and MIDI, we aim to extend such approaches to the under-studied medium of guitar tablature. Specifically, we develop the first work to our knowledge that models one specific genre as guitar tablature: heavy rock. Unlike other works in guitar tablature generation, we have a freely available public demo at https://huggingface.co/spaces/josuelmet/Metal_Music_Interpolator

Explore related subjects

Keep this discovery

BibTeXRIS

Josue Casco-Rodriguez. 2023-01-12. Rock Guitar Tablature Generation via Natural Language Processing. https://arxiv.org/abs/2301.05295

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Diarization Error Decomposition Under Pause Annotation Ambiguity

Speaker diarization evaluation is sensitive to ambiguity in pause annotation, which can inflate diarization error rate (DER) or obscure genuine model errors. We show that morphological closing, which has been used for pause-tolerant diarization evaluation, discards segment-level distinctions. Instead, we propose an exact, overlap-aware decomposition of standard DER into a pause-attributable component, consisting of errors compatible with pause filling, and a residual core component that can serve as a proxy for intrinsic diarization errors. The decomposition leaves DER unchanged, while the pause-attributable and core components vary monotonically with the pause threshold and eventually saturate. Experiments spanning synthetic transformations, annotation mismatch, cross-domain evaluation, and tight-boundary diarization show that the decomposition reveals error sources not apparent from standard DER.

eess.AS

Less can be More: What Aspects of Speech Drive End-of-Turn Detection

In conversational AI, detecting when a speaker has finished talking is crucial for natural turn taking. While recent work incorporates semantics, the relative contribution of different modalities remains unclear. We present a controlled ablation of acoustic, prosodic, and semantic signals for streaming end of turn detection using a lightweight trimodal classifier. Under identical training conditions, the acoustic prosodic combination achieves the best balance of accuracy and latency, achieving utterance F1 of 0.93 with 7.8% false alarms at 400ms median latency. Adding text increases premature detections without improving performance. Feature space analysis confirms that prosodic features have the strongest class separability, while text representations overlap substantially. These findings suggest that turn-taking is primarily conveyed through intonation and silence patterns rather than semantic completeness, enabling faster and more reliable systems without expensive text inference.

eess.AS

Downstream-Task-Aware Unified Source Separation

Task-aware unified source separation (TUSS) enables a single model to handle diverse separation tasks by conditioning on input prompts. However, conventional TUSS does not account for downstream task requirements, such as whether the enhanced speech will be used for human listening or automatic speech recognition (ASR). In this paper, we propose a prompt extension framework for TUSS that incorporates downstream task information into the input prompts and switches the loss function according to the given prompt during training, enabling outputs with different signal characteristics at inference time. Specifically, we introduce an ASR-dedicated prompt paired with a regularized loss function that reduces speech artifacts to improve ASR robustness, while the standard prompt is paired with the conventional SNR loss function. Experiments on the LibriSpeech and JNAS corpora demonstrate that the proposed joint-training scheme enables a single model to improve ASR performance over noisy input across a wide range of SNR conditions by selecting the ASR-dedicated prompt, while maintaining general speech enhancement quality when the standard prompt is used.

eess.AS