Local Speech Synthesis connector.

Generate short English speech locally and retrieve reusable WAV assets through an authenticated controller. List voices, submit, retrieve and cancel synthesis jobs.

Download connector See all connectors

Connector download

Download Local Speech Synthesis

Versioned ZIP for your CXFabric tenant.

Requires a separately deployed controller. Download controller source and setup instructions after signing in. Import only the connector ZIP into your tenant.

Use this connector in a flow

Register the ZIP in your tenant, or add a new version if the connector is already installed.
Create or select a credential if the action requires one.
In a Draft flow, add an available action, choose its version, connect it, and map inputs.
Test the step and flow; review output and failures before activation.

See readme.md in the ZIP for settings and limitations.

Version: 0.1.0.0
Release: 2026-10-17
Package: Implemented source
Schema credentials: Bearer token
Connection settings: Controller URL
Package size: 12 KB
SHA-256: e9a8e2c8ef3dcf96c431bb0cbd0015c761722ec50aa6b9587defad05cfd15831
Notes: Read release notes

Included in this version

Included actions

Implemented in v0.1.0.0. Use a separate flow trigger; no webhook trigger is included.

List Voices

List configured local voices and the fixed PCM WAV output format.

Submit Synthesis

Generate a short utterance locally. Completed jobs return an asset ID compatible with Play Audio.

Get Synthesis

Read job status and completed audio metadata. Failed or cancelled jobs return an error.

Cancel Synthesis

Stop an owned running synthesis job. Publication already in progress cannot be cancelled.

Vendor API

API and access

Vendor API documentation Vendor website

Implementation considerations

List Voices: List configured local voices and the fixed PCM WAV output format.
Submit Synthesis: Generate a short utterance locally. Completed jobs return an asset ID compatible with Play Audio.
Get Synthesis: Read job status and completed audio metadata. Failed or cancelled jobs return an error.
Cancel Synthesis: Stop an owned running synthesis job. Publication already in progress cannot be cancelled.
Deploy CXFabric FreeSWITCH Controller 0.3.0 and its optional pinned Kokoro-ONNX environment. Download controller source and setup separately from this page. Stage local read-only model and voice files with expected checksums; no hosted service, account, infrastructure or weights are supplied.
Use a controller-issued Bearer Token credential and trusted Controller URL from Flow Properties. List Voices returns configured English US/UK IDs. Supply short plain text up to 500 UTF-8 bytes and speed 0.75–1.5; output must fit 15 seconds and is not truncated. Output is 8 kHz mono 16-bit PCM WAV.
Retain the Request ID before submission. Read job status in later steps; only succeeded jobs without error have playable audio. Map assetId to the existing FreeSWITCH Call Control Play Audio action. No automatic polling, retries or redirects. Reusable cached audio stays within the same credential and exact input/model/voice identity.
One synthesis worker runs at a time with a separate 10–300 second processing deadline. Cancel Synthesis stops an owned running worker; once asset publication begins, retrieve its status instead. Restarted unfinished jobs fail without automatic resubmission. Job and asset quotas require operator retention.
Local tests use synthetic model/audio output over real HTTPS and SQLite. Hosted import, actual Kokoro inference, the optional synthesis image, real FreeSWITCH/carrier audio and intelligibility are unvalidated. File checks do not verify speech quality. This complete-prompt adapter does not implement streaming AI conversation or LiveKit bridging.

Support

Need help with Local Speech Synthesis?

Tell us what you need to connect or change.