Microsoft AI / NaturalSpeech 3
Factorized diffusion speech research system that separates content, prosody, timbre and acoustic detail. The scaled experiment uses one billion diffusion-model parameters and 200,000 hours of internal data.
Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.
No public service contract or verified value for this original research checkpoint.
No public service contract or verified value for this original research checkpoint.
No public service contract or verified value for this original research checkpoint.
Section 4.4 model-scale label; not a verified total including every codec and auxiliary component.
No public service contract or verified value for this original research checkpoint.
No public service contract or verified value for this original research checkpoint.
No public service contract or verified value for this original research checkpoint.
No public service contract or verified value for this original research checkpoint.
No public service contract or verified value for this original research checkpoint.
No public service contract or verified value for this original research checkpoint.
No public service contract or verified value for this original research checkpoint.
No public service contract or verified value for this original research checkpoint.
No public service contract or verified value for this original research checkpoint.
The 200K-hour internal training corpus does not provide a complete supported-language list. English evaluation does not establish an exhaustive inventory.
First research-family disclosure on arXiv; reviewed specifications may reflect a later paper revision. Not a commercial or separate variant release date.
Zero-shot research use of a reference voice, not an available commercial API.
An evaluation setting, not an advertised maximum reference duration.
Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.
Not collectedNo public API model selector.
Endpoint pricing has not been collected. Check the provider for applicable rates.
No hosted customer inference or data-retention contract verified.
Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.
The system processes eligible input and retains reusable state.
Reuse depends on matching input, session state and the cache policy.
Reusing state can reduce repeated work. Any charges follow the billing rules below.
No public service cache contract verified.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 14 Oct 2026No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.
Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.
Classification is an evidence-based inference for the original research checkpoint: the authors provide papers and prerecorded samples, but no full original model weights or callable service in the reviewed project materials. The Microsoft Hugging Face account lookup also returns no matching model. This does not classify independent implementations or separately released codec components as closed.
Classification evidence (opens in a new tab)