Pith. sign in

REVIEW 5 cited by

Robust Zero-Shot Text-to-Speech Synthesis with Reverse Inference Optimization

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 2407.02243 v1 pith:YYY4FODZ submitted 2024-07-02 cs.CL cs.SDeess.AS

classification cs.CLcs.SDeess.AS
keywords inferencereversespeechoptimizationrobustnesszero-shotgeneratedhuman
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

In this paper, we propose reverse inference optimization (RIO), a simple and effective method designed to enhance the robustness of autoregressive-model-based zero-shot text-to-speech (TTS) systems using reinforcement learning from human feedback (RLHF). To assess the quality of speech produced by the TTS system without human annotations, RIO introduces a novel concept termed as reverse inference based on the Bayesian principle, which suggests that a high-quality generated speech should be able to be used as a prompt for subsequent generation using the same TTS model. By leveraging reverse inference as the standard to select exemplars used in RLHF from the speech samples generated by the TTS system itself, RIO steers the subsequent optimization towards a direction of enhancing the TTS robustness. The RIO framework, comprising sampling, automatic annotating, and learning, obviates the need for a reward model or pairwise preference data, and significantly improves the stability of zero-shot TTS performance by reducing the discrepancies between training and inference conditions. Our experimental results verify that RIO can effectively improve both subjective and objective metrics, including mean opinion scores, word error rates, and speaker similarity. Remarkably, RIO can also diminish the incidence of bad outputs to nearly zero percent, rivalling the robustness when using ground-truth speech as the prompt.

Discussion (0). Sign in to comment.

Forward citations

Cited by 5 Pith papers

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

  1. Best-of-$N$ TTS Evaluation is Confounded by ASR Family Alignment

    cs.CL 2026-07 conditional novelty 6.0 of 10

    BoN TTS verifier rankings reverse across ASR families; same-family pairs recover 2–3× more oracle headroom, and cross-family rank ensembles give the most robust WER gains.

  2. MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech

    eess.AS 2025-08 conditional novelty 6.0 of 10

    MPO improves TTS alignment by constructing multi-dimensional preference pairs and adding cross-entropy regularization to DPO, yielding better intelligibility, speaker similarity, and prosody.

  3. Differentiable Reward Optimization for LLM based TTS system

    cs.SD 2025-07 conditional novelty 6.0 of 10

    DiffRO optimizes codec-based TTS models directly on differentiable token-level rewards, improving WER and enabling zero-shot emotion control.

  4. Robust and Efficient Autoregressive Speech Synthesis with Dynamic Chunk-wise Prediction Policy

    cs.SD 2025-06 conditional novelty 6.0 of 10

    DCAR dynamically schedules chunk-wise token prediction in AR TTS, improving WER by up to 72.27% relative and speeding up inference by up to 2.89x over next-token baselines.

  5. VoiceStar: Robust Zero-Shot Autoregressive TTS with Duration Control and Extrapolation

    eess.AS 2025-05 conditional novelty 6.0 of 10

    VoiceStar uses a progress-based rotary position embedding and mixed prompt training to give zero-shot voice cloning precise duration control and much longer output than training clips.

Pith tools