# Llasa: Scaling Train-Time and Inference-Time Compute for Llama-based Speech Synthesis

arXiv: https://arxiv.org/abs/2502.04128 · HTML: https://mlsanity.com/paper/2502.04128/

## Key facts

- Cited by 70 papers in the map (data through 14 Oct 2026).
- Placed in Audio-language model reasoning, one of 17 directions in LLM reasoning.
- Builds on 38 of its 65 references that are in the map.
- First posted on arXiv 6 Feb 2025.

## Placed in

- LLM reasoning / [Audio-language model reasoning](https://mlsanity.com/lr/audio-language-model-reasoning/)

## Builds on (most cited first, 30 of 38)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [OpenAI o1 System Card](https://mlsanity.com/paper/2412.16720/) (2024)
- [Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers](https://mlsanity.com/paper/2301.02111/) (2023)
- [FastSpeech 2: Fast and High-Quality End-to-End Text to Speech](https://mlsanity.com/paper/2006.04558/) (2020)
- [High Fidelity Neural Audio Compression](https://mlsanity.com/paper/2210.13438/) (2022)
- [Moshi: a speech-text foundation model for real-time dialogue](https://mlsanity.com/paper/2410.00037/) (2024)
- [MLS: A Large-Scale Multilingual Dataset for Speech Research](https://mlsanity.com/paper/2012.03411/) (2020)
- [GigaSpeech: An Evolving, Multi-domain ASR Corpus with 10,000 Hours of Transcribed Audio](https://mlsanity.com/paper/2106.06909/) (2021)
- [UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022](https://mlsanity.com/paper/2204.02152/) (2022)
- [F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching](https://mlsanity.com/paper/2410.06885/) (2024)
- [CosyVoice: A Scalable Multilingual Zero-shot Text-to-speech Synthesizer based on Supervised Semantic Tokens](https://mlsanity.com/paper/2407.05407/) (2024)
- [Seed-TTS: A Family of High-Quality Versatile Speech Generation Models](https://mlsanity.com/paper/2406.02430/) (2024)
- [CosyVoice 2: Scalable Streaming Speech Synthesis with Large Language Models](https://mlsanity.com/paper/2412.10117/) (2024)
- [Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps](https://mlsanity.com/paper/2501.09732/) (2025)
- [emotion2vec: Self-Supervised Pre-Training for Speech Emotion Representation](https://mlsanity.com/paper/2312.15185/) (2023)
- [Seamless: Multilingual Expressive and Streaming Speech Translation](https://mlsanity.com/paper/2312.05187/) (2023)
- [MaskGCT: Zero-Shot Text-to-Speech with Masked Generative Codec Transformer](https://mlsanity.com/paper/2409.00750/) (2024)
- [Paraformer: Fast and Accurate Parallel Transformer for Non-autoregressive End-to-End Speech Recognition](https://mlsanity.com/paper/2206.08317/) (2022)
- [WavTokenizer: an Efficient Acoustic Discrete Codec Tokenizer for Audio Language Modeling](https://mlsanity.com/paper/2408.16532/) (2024)
- [VALL-E 2: Neural Codec Language Models are Human Parity Zero-Shot Text to Speech Synthesizers](https://mlsanity.com/paper/2406.05370/) (2024)
- [FunASR: A Fundamental End-to-End Speech Recognition Toolkit](https://mlsanity.com/paper/2305.11013/) (2023)
- [BASE TTS: Lessons from building a billion-parameter Text-to-Speech model on 100K hours of data](https://mlsanity.com/paper/2402.08093/) (2024)
- [Codec Does Matter: Exploring the Semantic Shortcoming of Codec for Audio Language Model](https://mlsanity.com/paper/2408.17175/) (2024)
- [FireRedTTS: A Foundation Text-To-Speech Framework for Industry-Level Generative Speech Applications](https://mlsanity.com/paper/2409.03283/) (2024)
- [Autoregressive Speech Synthesis without Vector Quantization](https://mlsanity.com/paper/2407.08551/) (2024)
- [ELLA-V: Stable Neural Codec Language Modeling with Alignment-guided Sequence Reordering](https://mlsanity.com/paper/2401.07333/) (2024)

## Built on by (most cited first, 30 of 70)

- [Kimi-Audio Technical Report](https://mlsanity.com/paper/2504.18425/) (2025)
- [Qwen3-TTS Technical Report](https://mlsanity.com/paper/2601.15621/) (2026)
- [MiMo-Audio: Audio Language Models are Few-Shot Learners](https://mlsanity.com/paper/2512.23808/) (2025)
- [Recent Advances in Discrete Speech Tokens: A Review](https://mlsanity.com/paper/2502.06490/) (2025)
- [VibeVoice Technical Report](https://mlsanity.com/paper/2508.19205/) (2025)
- [OmniVoice: Towards Omnilingual Zero-Shot Text-to-Speech with Diffusion Language Models](https://mlsanity.com/paper/2604.00688/) (2026)
- [A Survey of Slow Thinking-based Reasoning LLMs using Reinforced Learning and Inference-time Scaling Law](https://mlsanity.com/paper/2505.02665/) (2025)
- [MOSS-TTS Technical Report](https://mlsanity.com/paper/2603.18090/) (2026)
- [TaDiCodec: Text-aware Diffusion Speech Tokenizer for Speech Language Modeling](https://mlsanity.com/paper/2508.16790/) (2025)
- [CapSpeech: Enabling Downstream Applications in Style-Captioned Text-to-Speech](https://mlsanity.com/paper/2506.02863/) (2025)
- [Towards Interactive Intelligence for Digital Humans](https://mlsanity.com/paper/2512.13674/) (2025)
- [FocalCodec-Stream: Streaming Low-Bitrate Speech Coding via Causal Distillation](https://mlsanity.com/paper/2509.16195/) (2025)
- [DMOSpeech 2: Reinforcement Learning for Duration Prediction in Metric-Optimized Speech Synthesis](https://mlsanity.com/paper/2507.14988/) (2025)
- [WenetSpeech-Chuan: A Large-Scale Sichuanese Corpus with Rich Annotation for Dialectal Speech Processing](https://mlsanity.com/paper/2509.18004/) (2025)
- [MagiCodec: Simple Masked Gaussian-Injected Codec for High-Fidelity Reconstruction and Generation](https://mlsanity.com/paper/2506.00385/) (2025)
- [J1: Exploring Simple Test-Time Scaling for LLM-as-a-Judge](https://mlsanity.com/paper/2505.11875/) (2025)
- [VoXtream: Full-Stream Text-to-Speech with Extremely Low Latency](https://mlsanity.com/paper/2509.15969/) (2025)
- [Dynin-Omni: Omnimodal Unified Large Diffusion Language Model](https://mlsanity.com/paper/2604.00007/) (2026)
- [FlowTTS-GRPO: Online Reinforcement Learning with Multi-Objective Reward Optimization for Flow-Matching Based Text-to-Speech](https://mlsanity.com/paper/2606.23190/) (2026)
- [Prompt-Unseen-Emotion: Zero-shot Expressive Speech Synthesis with Prompt-LLM Contextual Knowledge for Mixed Emotions](https://mlsanity.com/paper/2506.02742/) (2025)
- [No Verifiable Reward for Prosody: Toward Preference-Guided Prosody Learning in TTS](https://mlsanity.com/paper/2509.18531/) (2025)
- [UniSE: A Unified Framework for Decoder-Only Autoregressive LM-Based Speech Enhancement](https://mlsanity.com/paper/2510.20441/) (2025)
- [Beyond Fixed Frames: Dynamic Character-Aligned Speech Tokenization](https://mlsanity.com/paper/2601.23174/) (2026)
- [dots.tts Technical Report](https://mlsanity.com/paper/2606.07080/) (2026)
- [TTS-1 Technical Report](https://mlsanity.com/paper/2507.21138/) (2025)
- [Inference-time Scaling for Diffusion-based Audio Super-resolution](https://mlsanity.com/paper/2508.02391/) (2025)
- [SenSE: Semantic-Aware High-Fidelity Universal Speech Enhancement](https://mlsanity.com/paper/2509.24708/) (2025)
- [MOSS-Speech: Towards True Speech-to-Speech Models Without Text Guidance](https://mlsanity.com/paper/2510.00499/) (2025)
- [Multi-Reward GRPO for Stable and Prosodic Single-Codebook TTS LLMs at Scale](https://mlsanity.com/paper/2511.21270/) (2025)
- [Sparse Autoencoders for Interpretable Emotion Control in Text-to-Speech](https://mlsanity.com/paper/2606.01479/) (2026)

## Abstract (arXiv)

Recent advances in text-based large language models (LLMs), particularly in the GPT series and the o1 model, have demonstrated the effectiveness of scaling both training-time and inference-time compute. However, current state-of-the-art TTS systems leveraging LLMs are often multi-stage, requiring separate models (e.g., diffusion models after LLM), complicating the decision of whether to scale a particular model during training or testing. This work makes the following contributions: First, we explore the scaling of train-time and inference-time compute for speech synthesis. Second, we propose a simple framework Llasa for speech synthesis that employs a single-layer vector quantizer (VQ) codec and a single Transformer architecture to fully align with standard LLMs such as Llama. Our experiments reveal that scaling train-time compute for Llasa consistently improves the naturalness of synthesized speech and enables the generation of more complex and accurate prosody patterns. Furthermore, from the perspective of scaling inference-time compute, we employ speech understanding models as verifiers during the search, finding that scaling inference-time compute shifts the sampling modes toward the preferences of specific verifiers, thereby improving emotional expressiveness, timbre consistency, and content accuracy. In addition, we released the checkpoint and training code for our TTS model (1B, 3B, 8B) and codec model publicly available.