Skip to content

Configuration

cc-vox is configured via a TOML file at ~/.claude/cc-vox.toml. The file is created automatically on first run with sensible defaults.

Config File

[core]
enabled = true
voice = "af_heart"       # see voices below
backend = "auto"         # auto | kokoro | fish-speech | pocket-tts

[tuning]
speed = 1.0              # 0.5-2.0 (kokoro only)
max_sentences = 2        # max sentences in spoken summary (1-10)
fallback = true          # try other backends when forced one is down

[style]
prompt = ""              # custom voice personality instruction

Settings Reference

[core] Section

Key Type Default Description
enabled bool true Enable/disable voice feedback globally
voice string "af_heart" Voice name — Kokoro or pocket-tts alias
backend string "auto" TTS backend: auto, kokoro, fish-speech, pocket-tts

[tuning] Section

Key Type Default Range Description
speed float 1.0 0.5--2.0 Speech speed (Kokoro only)
max_sentences int 2 1--10 Maximum sentences in spoken summary
fallback bool true Try other backends if forced one is down

[style] Section

Key Type Default Description
prompt string "" Custom voice personality instruction

Voice personality examples

[style]
prompt = "be upbeat and encouraging"
[style]
prompt = "speak like a pirate"
[style]
prompt = "be concise and technical"

Modifying Config

/voice:speak backend kokoro    # Set backend
/voice:speak speed 1.3         # Set speed
/voice:speak prompt be chill   # Set personality
/voice:speak max_sentences 4   # Longer summaries

Via Direct File Edit

Edit ~/.claude/cc-vox.toml in any text editor. Changes take effect on the next Claude response.

Via Python API

from voice_common import update_voice_config

update_voice_config(backend="kokoro", speed=1.3)

Environment Variable Overrides

Environment variables take precedence over config file values for backend selection and port configuration.

Variable Default Description
TTS_BACKEND Override backend preference
KOKORO_PORT 32612 Kokoro Docker port
FISH_SPEECH_PORT 32611 Fish Speech Docker port
TTS_PORT 8000 pocket-tts port
GPU_THRESHOLD 80 GPU % above which Fish Speech is skipped
# Example: force Kokoro on a custom port
KOKORO_PORT=9000 TTS_BACKEND=kokoro claude

Migration

If you previously used the old voice.local.md config format, cc-vox automatically migrates your settings to TOML on first run. The old file is deleted after migration.