Input billingNot collectedAltered custom voice service
Output billingNot collectedNo inferred price
Specifications
Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.
Context / input limit
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Maximum input tokens
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Maximum output
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Parameters
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Release date
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Knowledge cutoff
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Tool calling
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Structured output
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Streaming
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Batch processing
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Fine-tuning
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Reasoning
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Context caching
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Base model
Not collected
No model-specific contract verified. Unknown does not mean unsupported.
Model terms
Altered licenses its application and voices without providing source-access rights; reverse engineering and unauthorized model redistribution are restricted. Free outputs require attribution and noncommercial use. Creator permits eligible individuals/companies of up to five employees with relevant trailing-year revenue/funding below USD 100,000. Professional uses the stated commercial license. Custom voice creation requires source rights and speaker consent; continued use requires a valid subscription. No public model weights verified.Source for Model terms (opens in a new tab)
Creation privacy
Generic terms permit content/output collection for diagnostics and service improvement and reserve a private record of source-voice files. Privacy-policy retention is purpose/necessity-based, not a fixed inference TTL. Local-cloning and Standard Real-Time Pro pages make narrower local-processing promises; discounted realtime plans explicitly permit anonymized voice data for improvement. Exact precedence, contractual exceptions and cloud handling remain unresolved.Source for Creation privacy (opens in a new tab)
Service suggests at least30 minutes clean recordings; FAQ gives60–90 minutes, sometimes30 minutes depending on voice coverage, and shorter specialized work at extra cost. These are guidance, not a single universal hard minimum.Source for Reference audio (opens in a new tab)
Model selection
Provider prepares a private customer voice key for its tools. Custom models can support TTS or STS; exact base checkpoint, training method and individual model inventory are private.Source for Model selection (opens in a new tab)
Published price
Not collected
Contact-sales custom quote; no public numeric per-voice price verified.
Access & pricing
Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.
No callable public model ID or immutable checkpoint selector verified; UI names and voice-profile IDs are not API contracts.
Access
Contact-based custom voice engagement. Reference rights and speaker consent required; provider-issued private key unlocks the voice in its tools. No public self-service training endpoint verified.
Regions
Processing residency depends on deployment; no universal model-specific region guarantee verified.
Endpoint pricing has not been collected. Check the provider for applicable rates.
Data & enterprise terms
Generic terms permit content/output collection for diagnostics and service improvement and reserve a private record of source-voice files. Privacy-policy retention is purpose/necessity-based, not a fixed inference TTL. Local-cloning and Standard Real-Time Pro pages make narrower local-processing promises; discounted realtime plans explicitly permit anonymized voice data for improvement. Exact precedence, contractual exceptions and cloud handling remain unresolved.
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.
Altered custom voice service
Not verified
No inference prompt-cache contract verified. Repeating a morph of the same input without additional quota, stored voice profiles and local processing are not token caching.
How it works
Not collected
No inference prompt-cache contract verified. Repeating a morph of the same input without additional quota, stored voice profiles and local processing are not token caching.
Minimum reusable input
Not collected
No inference prompt-cache contract verified. Repeating a morph of the same input without additional quota, stored voice profiles and local processing are not token caching.
Lifetime & refresh
Not collected
No inference prompt-cache contract verified. Repeating a morph of the same input without additional quota, stored voice profiles and local processing are not token caching.
How to enable it
Not collected
No inference prompt-cache contract verified. Repeating a morph of the same input without additional quota, stored voice profiles and local processing are not token caching.
Usage fields
Not collected
No inference prompt-cache contract verified. Repeating a morph of the same input without additional quota, stored voice profiles and local processing are not token caching.
Billing rules
Not collected
No inference prompt-cache contract verified. Repeating a morph of the same input without additional quota, stored voice profiles and local processing are not token caching.
Limits & data boundaries
Not collected
No inference prompt-cache contract verified. Repeating a morph of the same input without additional quota, stored voice profiles and local processing are not token caching.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 17 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
Voice enrollment creates a customer adaptation/profile using Altered technology. This workflow is explicitly managed; each customer voice is not a newly enumerated native base checkpoint.
