voice/clone

Voice Cloning workflow — clone voices from audio samples.

Category: voice
Source: workflows/voice/clone.py

Input Schema

Field	Type	Default	Description
`description`	`string`	`""`	Voice characteristics description
`persist`	`boolean`	`false`	Save for reuse across runs
`provider`	`string`	`"auto"`	Provider: ‘elevenlabs’, ‘fal-chatterbox’, or ‘auto’
`regenerate`	`object`	—	When set, this run is a regeneration. Workflows may read `direction` / `keep` / `extra_instructions` to modulate prompts; the engine persists `parent_run_id` and `parent_variant_index` as run lineage columns.
`sample_paths`	`string[]`	—	Audio file paths (1+ samples)
`sample_urls`	`string[]`	—	Audio URLs (alternative to paths)
`variants`	`integer`	`1`	Number of independent variant executions (1–10). When > 1, the engine runs the workflow N times with different sampling, producing N outputs.
`voice_name`	`string`	`"cloned-voice"`	Human-readable name for the clone

Output Schema

Field	Type	Default	Description
`kind`	`object`	—	Variant card shape: video / carousel / image / text. Surfaced on the per-variant entry of the run-output API and used by gallery UIs to pick the right layout.
`persisted`	`boolean`	`false`	Whether the voice was saved for reuse
`provider`	`string`	`""`	Provider that created the clone
`voice_id`	`string`	`""`	Provider-assigned voice ID
`voice_name`	`string`	`""`	Name given to the cloned voice

Task Pipeline

clone_voice

Task	Description
`clone_voice`	Clone a voice — delegates to SDK stage.

Run-spec example

Save the YAML below as my-run.yaml, edit the values, and run with the CLI or POST it to the API. Required fields are uncommented; optional knobs are documented above the input: block — copy any line under input: and uncomment to set.

workflow: voice/clone

# Optional fields — copy any line(s) under `input:` and uncomment to set:
# Voice characteristics description
#   description: ""
#
# Save for reuse across runs
#   persist: false
#
# Provider: 'elevenlabs', 'fal-chatterbox', or 'auto'
#   provider: auto
#
# Audio file paths (1+ samples)
#   sample_paths: []
#
# Audio URLs (alternative to paths)
#   sample_urls: []
#
# Human-readable name for the clone
#   voice_name: cloned-voice
#

input: {}

Run it locally:

fab-workflow --from-file my-run.yaml

Or submit over the wire — the same file is the request body:

curl -X POST 'https://gofabric.dev/v1/workflows/runs?name=voice/clone' \
  -H 'Authorization: Bearer fab_xxx' \
  -H 'content-type: application/yaml' \
  --data-binary @my-run.yaml

Every workflow also accepts the universal WorkflowInput fields — variants (1–10 fan-out) and regenerate (creative-direction hints with run lineage). See Run-specs (YAML / TOML / JSON) for the full top-level shape (metadata, priority, bundle, parent, etc.).

Warnings

Task clone_voice has no Pydantic types — contract is opaque to consumers.