voice — Help
Generated from voice --help and voice <cmd> --help (v3.1, 14 commands).
voice --help
usage: voice [-h] [-v] [--non-interactive] [--version]
[--completion {bash,zsh,fish}]
{get,cast,delta,graft,chorus,add,list,info,remove,cleanup,auth,report,path,doctor} ...
Lift a model's voice and cast it onto another model.
positional arguments:
{get,cast,delta,graft,chorus,add,list,info,remove,cleanup,auth,report,path,doctor}
get Get a voice from a model (safetensors repo, .gguf, or
URL)
cast Cast a voice onto a .gguf or .safetensors file
delta Make a reusable style (voice − base)
graft Graft one voice onto another (DELLA single graft)
chorus Blend voices: pulse + support + lead (DELLA chorus)
add Add a voice from local voice.safetensors + voice.json
list List stored voices
info Show voice details
remove Remove a stored voice
cleanup Remove incomplete downloads
auth Save your HF token (optional)
report Save a report locally and optionally share to
community
path Put voice on your PATH
doctor Check your setup
options:
-h, --help show this help message and exit
-v, --verbose Verbose: tensor shapes, decisions, and paths
--non-interactive Fail instead of prompting (same as piped stdin; needs
--yes to overwrite/remove)
--version Show version and environment
--completion {bash,zsh,fish}
Print shell completion script
voice get
usage: voice get [-h] [-q] [--name NAME] [--multi] [--out OUT] [--yes]
source [tensors ...]
positional arguments:
source HF model ID, .gguf/.safetensors path, or URL
tensors Tensor name(s) to fetch (default: auto-detect output head; 2+
or --multi saves a voicepack)
options:
-h, --help show this help message and exit
-q, --quiet Quieter: progress and success notes off (warnings and errors
still print)
--name NAME Voice name to save (registry voices only; not with --out)
--multi Fetch the named tensors (1+) as voicepack.safetensors +
voicepack.json (not with --name)
--out OUT Write a standalone file instead of a registry voice (default
pack name: voicepack.safetensors)
--yes Assume yes to prompts (overwrite without asking)
voice cast
usage: voice cast [-h] [--out OUT] [--yes] [--quant QUANT]
[--exclude EXCLUDES] [--speak] [-q]
target voice [tensors ...]
positional arguments:
target Path to the target .gguf or .safetensors
voice Voice name, or a .safetensors/.gguf path
tensors Only cast these tensors (default: every tensor in the
voice)
options:
-h, --help show this help message and exit
--out OUT Output file or folder (default:
~/Voiced/<voice>-<model>/<voice>-<model>.<ext matching
target>; a folder invents the filename)
--yes Assume yes to prompts (overwrite without asking)
--quant QUANT Output quant override: keep (default, preserve quality)
| target | any quant
(q8_0,q4_0,q4_k,q6_k,iq4_nl,f16,f32,bf16,...;
safetensors targets: keep|target|f32|f16|bf16)
--exclude EXCLUDES Skip tensors on cast (repeatable: --exclude foo
--exclude bar, or --exclude foo,bar)
--speak Voice speaks through output.weight (invented if absent
when head/embed geometry survives; GGUF targets only)
-q, --quiet Quieter: progress and success notes off (warnings and
errors still print)
voice delta
usage: voice delta [-h] [--base BASE] [--yes] [-q] [voicename]
positional arguments:
voicename Voice name (default: pick from registry)
options:
-h, --help show this help message and exit
--base BASE Base HF repo (org/model) or local path to base safetensors
--yes Assume yes to prompts (overwrite without asking)
-q, --quiet Quieter: progress and success notes off (warnings and errors
still print)
voice graft
usage: voice graft [-h] [--density DENSITY] [--epsilon EPSILON]
[--lambda_ LAMBDA_] [--rescale {l1,inv_p}] [--seed SEED]
[--out OUT] [--yes] [-q]
[pulse] [donor]
positional arguments:
pulse Pulse voice name (default: pick from registry)
donor Donor voice name (default: pick from registry)
options:
-h, --help show this help message and exit
--density DENSITY Keep-fraction for magnitude pruning (default: 0.8)
--epsilon EPSILON Rank-ramp half-width (default: 0.1)
--lambda_ LAMBDA_ Global rescale after combine (default: 1.0)
--rescale {l1,inv_p} Prune rescale: l1 (stable) or inv_p (paper-faithful)
--seed SEED Bernoulli seed for reproducible grafts (default: 42)
--out OUT Output voice name (default: chorus-<pulse>-<donor>)
--yes Assume yes to prompts (overwrite without asking)
-q, --quiet Quieter: progress and success notes off (warnings and
errors still print)
voice chorus
usage: voice chorus [-h] [--config] [--density DENSITY]
[--lead-density LEAD_DENSITY] [--epsilon EPSILON]
[--lead-epsilon LEAD_EPSILON] [--lambda_ LAMBDA_]
[--rescale {l1,inv_p}] [--seed SEED] [--out OUT] [--yes]
[-q]
[pulse] [donors ...]
positional arguments:
pulse Pulse voice name (default: pick interactively)
donors Support + lead voice names (0-2; 1 = duet at lead
density)
options:
-h, --help show this help message and exit
--config Stage knob screens per graft (Enter accepts each
default; screens prefill from flags)
--density DENSITY First-graft density (default: 0.8)
--lead-density LEAD_DENSITY
Lead-graft density (default: 0.9)
--epsilon EPSILON Rank-ramp half-width (default: 0.1)
--lead-epsilon LEAD_EPSILON
Lead-graft epsilon (default: 0.05; keeps 0.9±eps
valid)
--lambda_ LAMBDA_ Global rescale after combine (default: 1.0)
--rescale {l1,inv_p} Prune rescale: l1 (stable) or inv_p (paper-faithful)
--seed SEED Bernoulli seed for reproducible choruses (default: 42)
--out OUT Final output voice name (default:
chorus-<pulse>-<lead>)
--yes Assume yes to prompts (overwrite without asking)
-q, --quiet Quieter: progress and success notes off (warnings and
errors still print)
voice add
usage: voice add [-h] [--name NAME] [--yes] [-q] safetensors json
positional arguments:
safetensors Path to voice.safetensors (required)
json Path to voice.json (required)
options:
-h, --help show this help message and exit
--name NAME Voice name to save (default from voice.json)
--yes Auto-confirm overwrite
-q, --quiet Quieter: progress and success notes off (warnings and errors
still print)
voice list
usage: voice list [-h] [--base] [-q] [base]
positional arguments:
base Show cached base heads (same as --base)
options:
-h, --help show this help message and exit
--base Show cached base heads instead of voices
-q, --quiet Quieter: progress and success notes off (warnings and errors
still print)
voice info
usage: voice info [-h] [-q] voicename
positional arguments:
voicename Voice name
options:
-h, --help show this help message and exit
-q, --quiet Quieter: progress and success notes off (warnings and errors
still print)
voice remove
usage: voice remove [-h] [--yes] [-q] voice
positional arguments:
voice Voice name
options:
-h, --help show this help message and exit
--yes Assume yes to prompts (remove without asking)
-q, --quiet Quieter: progress and success notes off (warnings and errors
still print)
voice cleanup
usage: voice cleanup [-h] [--force] [-q]
options:
-h, --help show this help message and exit
--force Also clear in-progress downloads (locked)
-q, --quiet Quieter: progress and success notes off (warnings and errors
still print)
voice auth
usage: voice auth [-h] [--clear] [--show] [-q]
options:
-h, --help show this help message and exit
--clear Remove the stored token
--show Show the masked token
-q, --quiet Quieter: progress and success notes off (warnings and errors
still print)
voice report
usage: voice report [-h] [--message MESSAGE] [--voice VOICE] [--target TARGET]
[--reporter REPORTER] [--yes] [--no-push] [--sync] [-q]
options:
-h, --help show this help message and exit
--message, -m MESSAGE
Freeform note (240ch, sanitized)
--voice VOICE Voice name context
--target TARGET Target file basename context
--reporter REPORTER Reporter id (optional, 64ch)
--yes Non-interactive: skip preview and auto-share to
community
--no-push Local only, skip share prompt
--sync Sync pending outbox to community
-q, --quiet Quieter: progress and success notes off (warnings and
errors still print)
voice path
usage: voice path [-h] [-q]
options:
-h, --help show this help message and exit
-q, --quiet Quieter: progress and success notes off (warnings and errors
still print)
voice doctor
usage: voice doctor [-h] [-q]
options:
-h, --help show this help message and exit
-q, --quiet Quieter: progress and success notes off (warnings and errors
still print)