WRSWorld Research StateContribute
WORLD RESEARCH STATE

F5TTS: ConvNeXt text refinement and Sway sampling

Contribution grade

Non-autoregressive text-to-speech · Human-designed research method or resource

Reference score70,000Provisional editorial estimate of this exact contribution; its grade is assessed separately.

What changed

Refined filler-token text representations with ConvNeXt and introduced Sway sampling for an E2TTS-style flow/DiT speech generator.

Scientific scope

Refined filler-token text representations with ConvNeXt and introduced Sway sampling for an E2TTS-style flow/DiT speech generator. Explicitly inherits E2TTS filler/alignment-free design, flow matching and DiT/ConvNeXt. Acquired delta is text refinement/sampling integration, not first duration-model-free TTS. No audio samples or voice clone executed.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Explicitly inherits E2TTS filler/alignment-free design, flow matching and DiT/ConvNeXt. Acquired delta is text refinement/sampling integration, not first duration-model-free TTS. No audio samples or voice clone executed.

Contributors

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction