用单流编码器实现高效可控语音合成，支持精细调节音调与风格。
Spark-TTS: An Efficient LLM-Based Text-to-Speech Model with Single-Stream Decoupled Speech Tokens
- 采用单流双类型令牌分离语义与说话人特征，提升生成效率。
- 在10万小时数据集上实现顶尖零样本语音克隆效果。
- 适合需要高定制化语音生成的研究者与开发者使用。
大型语言模型（LLM）的进展推动了零样本语音合成的重大突破。然而，现有基础模型依赖多阶段处理或复杂架构预测多个码本，限制了效率与集成灵活性。为此，我们提出Spark-TTS，基于BiCodec单流语音编码器，将语音分解为低比特率语义令牌（表征语言内容）与固定长度全局令牌（表征说话人属性）。该解耦表示结合Qwen2.5 LLM与链式思维（CoT）生成方法，可实现粗粒度控制（如性别、语调风格）与细粒度调整（如精确音高值、语速）。为促进可控语音合成研究，我们构建了VoxBox数据集，包含10万小时高质量语音及全面属性标注。大量实验证明，Spark-TTS不仅达到当前最优的零样本语音克隆性能，还能生成超越参考基合成限制的高度可定制语音。源代码、预训练模型与音频样例已开源：https://github.com/SparkAudio/Spark-TTS。
原文摘要 · Abstract (English)
Recent advancements in large language models (LLMs) have driven significant progress in zero-shot text-to-speech (TTS) synthesis. However, existing foundation models rely on multi-stage processing or complex architectures for predicting multiple codebooks, limiting efficiency and integration flexibility. To overcome these challenges, we introduce Spark-TTS, a novel system powered by BiCodec, a single-stream speech codec that decomposes speech into two complementary token types: low-bitrate semantic tokens for linguistic content and fixed-length global tokens for speaker attributes. This disentangled representation, combined with the Qwen2.5 LLM and a chain-of-thought (CoT) generation approach, enables both coarse-grained control (e.g., gender, speaking style) and fine-grained adjustments (e.g., precise pitch values, speaking rate). To facilitate research in controllable TTS, we introduce VoxBox, a meticulously curated 100,000-hour dataset with comprehensive attribute annotations. Extensive experiments demonstrate that Spark-TTS not only achieves state-of-the-art zero-shot voice cloning but also generates highly customizable voices that surpass the limitations of reference-based synthesis. Source code, pre-trained models, and audio samples are available at https://github.com/SparkAudio/Spark-TTS.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。