Pith. sign in

REVIEW 10 cited by

SongGen: A Single Stage Auto-regressive Transformer for Text-to-Song Generation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2502.13128 v2 pith:KZ23KDTP submitted 2025-02-18 cs.SD cs.AI

SongGen: A Single Stage Auto-regressive Transformer for Text-to-Song Generation

classification cs.SD cs.AI
keywords generationsonggenauto-regressivedatamodeaccompanimentcodecontrol
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Text-to-song generation, the task of creating vocals and accompaniment from textual inputs, poses significant challenges due to domain complexity and data scarcity. Existing approaches often employ multi-stage generation procedures, leading to cumbersome training and inference pipelines, as well as suboptimal overall generation quality due to error accumulation across stages. In this paper, we propose SongGen, a fully open-source, single-stage auto-regressive transformer designed for controllable song generation. The proposed model facilitates fine-grained control over diverse musical attributes, including lyrics and textual descriptions of instrumentation, genre, mood, and timbre, while also offering an optional three-second reference clip for voice cloning. Within a unified auto-regressive framework, SongGen supports two output modes: mixed mode, which generates a mixture of vocals and accompaniment directly, and dual-track mode, which synthesizes them separately for greater flexibility in downstream applications. We explore diverse token pattern strategies for each mode, leading to notable improvements and valuable insights. Furthermore, we design an automated data preprocessing pipeline with effective quality control. To foster community engagement and future research, we will release our model weights, training code, annotated data, and preprocessing pipeline. The code is available at https://github.com/LiuZH-19/SongGen.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 10 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Audio-Oscar: A Multi-Agent System for Complex Audio Scene Generation, Orchestration, and Refinement

    cs.SD 2026-06 unverdicted novelty 7.0

    Audio-Oscar is a multi-agent system that coordinates specialist agents for generating audio from complex scene descriptions and introduces the ASG-Bench benchmark for evaluation.

  2. MIDI-Informed Singing Accompaniment Generation in a Compositional Song Pipeline

    cs.SD 2026-02 unverdicted novelty 7.0

    MIDI-SAG generates consistent long-form singing accompaniments by feeding symbolic MIDI timing, chords, and structure labels into a compositional pipeline built from pre-trained modules.

  3. MusicMark: A Robust Generative Watermarking Framework for Music Generation

    cs.SD 2026-07 conditional novelty 6.5

    Embedding watermark bits into diffusion semantic latents via a frozen-backbone adapter yields far more robust music provenance than post-hoc watermarking under codecs and cover-song attacks.

  4. Qwen-Music Technical Report

    cs.SD 2026-07 conditional novelty 6.0

    Qwen-Music generates high-fidelity vocal songs via 25 Hz semantic tokens, Melody-CoT planning, and DiT rendering, claiming SOTA on 13/16 metrics and expert preference over proprietary systems.

  5. MMGenre: Benchmarking Singing Voice Synthesis across Multiple Musical Genres

    cs.SD 2026-07 conditional novelty 6.0

    Current singing voice synthesis models fail to differentiate musical genres, defaulting to pop-like output regardless of input genre, unless given genre-specific fine-tuning data.

  6. Unified Audio Intelligence Without Regressing on Text Intelligence

    cs.CL 2026-07 conditional novelty 6.0

    Audex unifies audio understanding and generation on a strong text MoE backbone with multi-stage SFT plus text-only Cascade RL, matching open SOTA audio scores while mostly retaining text capability.

  7. Real-Time Interactive Music Generation via Data-Free Streaming Consistency Distillation

    cs.SD 2026-06 unverdicted novelty 6.0

    A data-free streaming consistency distillation framework enables single-step autoregressive generation from text-to-music models for real-time interactive use while preserving timbre and rhythm via latent, spectral, a...

  8. UniVocal: Unified Speech-Singing Code-Switching Synthesis

    cs.SD 2026-06 unverdicted novelty 6.0

    UniVocal presents a text-context-only framework for speech-singing code-switching synthesis via two-stage curriculum learning and a synthetic data pipeline, claiming SOTA on a new benchmark.

  9. Unified Audio Intelligence Without Regressing on Text Intelligence

    cs.CL 2026-07 conditional novelty 5.0

    A unified 30B MoE audio-text LLM achieves state-of-the-art audio understanding, generation, and speech tasks while preserving text reasoning comparable to its text-only backbone.

  10. LeVo 2: Stable and Melodious Song Generation via Hierarchical Representation Modeling and Progressive Post-Training

    cs.SD 2026-06 unverdicted novelty 5.0

    LeVo 2 presents a hierarchical LLM-Diffusion model with progressive post-training stages to generate full-length songs that balance semantic planning, track-specific acoustics, and musicality.