Proprietary neural speech synthesis that generates waveforms, available through licensed CereProc deployments. This is a named model family, not an immutable checkpoint or a cloud model selector.
Custom deep neural network synthesizing speech waveforms; exact architecture, parameter count and checkpoint revisions are not specified.Source for Architecture (opens in a new tab)
Deployment options
Licensed SDK for Windows, macOS, Linux, Android and iOS; cServer for Windows/Linux; CerePlay mobile/offline deployment. SDK development and commercial redistribution require different licensing arrangements.Source for Deployment options (opens in a new tab)
CerePlay and cServer are separately documented products, not additional base models.
The same page assigns 8/48 kHz to Unit Selection voices. Do not copy those rates into the neural engine or treat resampled cloud output as a native model rate.
Voice controls
Provider describes adaptive voice construction, changing voice characteristics and expressive synthesis. This does not establish a public fine-tuning endpoint or universal per-voice controls.Source for Voice controls (opens in a new tab)
Stock voice catalog
Not collected
No authoritative complete CereWave model/voice/version mapping collected. The voice shop now publishes a desktop deprecation notice rather than a full inventory.
Published price
Not collected
Commercial SDK/server rates are arranged through sales; cloud character-pack pricing does not establish a CereWave-specific deployment tariff.
Documentation discrepancy
The shop says Windows SAPI5 and macOS voices are being deprecated while voice development continues through Capacity AI. This is not evidence that all CereWave models, cServer or Cloud are retired. Product availability and historical checkpoint revisions require separate confirmation.Source for Documentation discrepancy (opens in a new tab)
Model terms
Desktop voices have personal single-device licence terms. SDK developer licences allow integration/testing; commercial release requires an agreement through sales. Do not apply desktop terms as the complete commercial or cloud contract.Source for Model terms (opens in a new tab)
Access & pricing
Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.
No public cloud model selector; access is through the licensed product and voice build.
Access
Licensed SDK; annual developer access and separate commercial distribution terms. Documentation lists Windows, macOS, Linux, Android and iOS; exact runtime/model-build compatibility needs confirmation.
Regions
Local or customer-managed deployment; no provider-hosted region implied.
Endpoint pricing has not been collected. Check the provider for applicable rates.
Data & enterprise terms
Commercial deployment and redistribution require product-specific licensing. No current customer deployment data policy established by the reviewed marketing pages.
No public cloud model selector; access is through the licensed product and voice build.
Access
Licensed Windows/Linux synthesis server, cSpeech C/C++ and Python clients and SAPI 5 client. Neural voices use documented 16/24 kHz output; simultaneous streams do not establish a public streaming HTTP model endpoint.
Regions
Local or customer-managed deployment; no provider-hosted region implied.
Endpoint pricing has not been collected. Check the provider for applicable rates.
Data & enterprise terms
Commercial deployment and redistribution require product-specific licensing. No current customer deployment data policy established by the reviewed marketing pages.
No public cloud model selector; access is through the licensed product and voice build.
Access
Provider describes offline CereWave voices on Android, iOS and iPadOS through CerePlay. App/voice entitlement and actual model build depend on the installed product.
Regions
Local or customer-managed deployment; no provider-hosted region implied.
Endpoint pricing has not been collected. Check the provider for applicable rates.
Data & enterprise terms
Commercial deployment and redistribution require product-specific licensing. No current customer deployment data policy established by the reviewed marketing pages.
Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.
01 / INITIAL INPUTProcess & retain
The system processes eligible input and retains reusable state.
02 / LATER REQUESTCheck available state
Reuse depends on matching input, session state and the cache policy.
03 / CACHE HITReuse & generate
Reusing state can reduce repeated work. Any charges follow the billing rules below.
CereVoice SDK
Not verified
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
How it works
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Minimum reusable input
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Lifetime & refresh
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
How to enable it
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Usage fields
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Billing rules
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Limits & data boundaries
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 16 Oct 2026
CereVoice cServer
Not verified
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
How it works
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Minimum reusable input
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Lifetime & refresh
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
How to enable it
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Usage fields
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Billing rules
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Limits & data boundaries
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 16 Oct 2026
CerePlay
Not verified
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
How it works
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Minimum reusable input
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Lifetime & refresh
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
How to enable it
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Usage fields
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Billing rules
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
Limits & data boundaries
Not collected
No documented inference-cache contract. Saved audio, access tokens and stored lexicons are not prompt caching.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 16 Oct 2026
What the measurements say
No 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.
The record behind the model
Why it is included
CereProc retains proprietary software rights and licenses use rather than redistributing source or weights. Its voice and FAQ pages identify CereWave as the custom DNN waveform model. This family record does not classify the separate Unit Selection engine as a neural model.
