# DuoTok: Source-Aware Dual-Track Music Tokenization for Vocal-Accompaniment Generation

arXiv: https://arxiv.org/abs/2511.20224 · HTML: https://mlsanity.com/paper/2511.20224/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 30 references that are in the map.
- First posted on arXiv 25 Nov 2025.

## Builds on (most cited first, 23 of 23)

- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://mlsanity.com/paper/2110.13900/) (2021)
- [SoundStream: An End-to-End Neural Audio Codec](https://mlsanity.com/paper/2107.03312/) (2021)
- [High-Fidelity Audio Compression with Improved RVQGAN](https://mlsanity.com/paper/2306.06546/) (2023)
- [BEATs: Audio Pre-Training with Acoustic Tokenizers](https://mlsanity.com/paper/2212.09058/) (2022)
- [WhisperX: Time-Accurate Speech Transcription of Long-Form Audio](https://mlsanity.com/paper/2303.00747/) (2023)
- [MERT: Acoustic Music Understanding Model with Large-Scale Self-supervised Training](https://mlsanity.com/paper/2306.00107/) (2023)
- [WavTokenizer: an Efficient Acoustic Discrete Codec Tokenizer for Audio Language Modeling](https://mlsanity.com/paper/2408.16532/) (2024)
- [Music Source Separation in the Waveform Domain](https://mlsanity.com/paper/1911.13254/) (2019)
- [FunASR: A Fundamental End-to-End Speech Recognition Toolkit](https://mlsanity.com/paper/2305.11013/) (2023)
- [Addressing Representation Collapse in Vector Quantized Models with One Linear Layer](https://mlsanity.com/paper/2411.02038/) (2024)
- [Codec Does Matter: Exploring the Semantic Shortcoming of Codec for Audio Language Model](https://mlsanity.com/paper/2408.17175/) (2024)
- [YuE: Scaling Open Foundation Models for Long-Form Music Generation](https://mlsanity.com/paper/2503.08638/) (2025)
- [MuQ: Self-Supervised Music Representation Learning with Mel Residual Vector Quantization](https://mlsanity.com/paper/2501.01108/) (2025)
- [SemantiCodec: An Ultra Low Bitrate Semantic Audio Codec for General Sound](https://mlsanity.com/paper/2405.00233/) (2024)
- [Music Source Separation with Band-Split RoPE Transformer](https://mlsanity.com/paper/2309.02612/) (2023)
- [SongGen: A Single Stage Auto-regressive Transformer for Text-to-Song Generation](https://mlsanity.com/paper/2502.13128/) (2025)
- [XY-Tokenizer: Mitigating the Semantic-Acoustic Conflict in Low-Bitrate Speech Codecs](https://mlsanity.com/paper/2506.23325/) (2025)
- [Ming-UniAudio: Speech LLM for Joint Understanding, Generation and Editing with Unified Representation](https://mlsanity.com/paper/2511.05516/) (2025)
- [SongCreator: Lyrics-based Universal Song Generation](https://mlsanity.com/paper/2409.06029/) (2024)
- [MuCodec: Ultra Low-Bitrate Music Codec](https://mlsanity.com/paper/2409.13216/) (2024)
- [Optimized Self-supervised Training with BEST-RQ for Speech Recognition](https://mlsanity.com/paper/2501.16131/) (2025)

## Abstract (arXiv)

Multi-track music generation requires tokens that preserve acoustic fidelity, support sequence modeling, and maintain cross-track structure. Reconstruction-oriented codecs retain acoustic detail but are difficult to model, while semantic tokenizers may sacrifice fidelity or cross-track alignment. We present DuoTok, a source-aware dual-track music tokenizer for vocal-accompaniment generation based on staged disentanglement. DuoTok first learns a semantic audio representation through self-supervised pretraining, then shapes source-aware structure using feature replacement noise and multi-task supervision: spectral reconstruction, music source separation regularization, and an ASR head for lyric alignment. It freezes the encoder and learns hard-routed codebooks for vocals and accompaniment, while a diffusion decoder restores fine acoustic detail from discrete tokens. Across public benchmarks, DuoTok achieves a favorable predictability-fidelity trade-off at ultra-low bitrate. Under held-constant dual-track language modeling, it improves both unconditional vocal-accompaniment modeling and vocal-conditioned accompaniment prediction. Controlled diagnostics show larger predictability costs under cross-track corruption and greater gains from longer temporal context, supporting stronger use of cross-track and temporal structure rather than merely easier local prediction. DuoTok also maintains competitive reconstruction quality and preserves control-relevant musical attributes in its discrete space. These results suggest that tokenizer design is a core modeling problem for multi-track music generation, beyond compression alone.