refactor(hermes_cli): setup_tts — derive short labels from the choice table; collapse install error paths

This commit is contained in:
Teknium
2026-09-02 20:31:59 -07:00
parent 8f03ede4f9
commit c227ea7e1a
+36 -65
View File
@@ -1,6 +1,7 @@
"""Text-to-speech provider setup (provider picker, API-key prompts, local engine installs, xAI OAuth).
Extracted from hermes_cli/setup.py; setup.py re-exports the public entry points.
Names from setup.py are imported lazily inside bodies so test patches on
``hermes_cli.setup.<name>`` take effect; setup.py re-exports the public entry points.
"""
import logging
@@ -11,11 +12,6 @@ import sys
logger = logging.getLogger("hermes_cli.setup")
def _check_espeak_ng() -> bool:
"""Check if espeak-ng is installed."""
return shutil.which("espeak-ng") is not None or shutil.which("espeak") is not None
def _pip_install_tts_package(name: str, pip_args: list, manual_cmd: str) -> bool:
"""Install a local TTS engine through the canonical uv → pip → ensurepip
ladder so pip-less venvs (Ubuntu 25.10 ``python -m venv``, ``uv venv``) work."""
@@ -23,15 +19,14 @@ def _pip_install_tts_package(name: str, pip_args: list, manual_cmd: str) -> bool
from hermes_cli.tools_config import _pip_install
try:
result = _pip_install(pip_args, timeout=300)
if result.returncode == 0:
print_success(f"{name} installed successfully")
return True
err = (result.stderr or "").strip()
reason = err[:300] if err else "install failed"
except Exception as e:
print_error(f"Failed to install {name}: {e}")
print_info(f"Try manually: {manual_cmd}")
return False
if result.returncode == 0:
print_success(f"{name} installed successfully")
return True
err = (result.stderr or "").strip()
print_error(f"Failed to install {name}: {err[:300] if err else 'install failed'}")
reason = e
print_error(f"Failed to install {name}: {reason}")
print_info(f"Try manually: {manual_cmd}")
return False
@@ -47,7 +42,7 @@ _ESPEAK_INSTALL = {
def _install_neutts_deps() -> bool:
"""Install NeuTTS dependencies with user approval. Returns True on success."""
from hermes_cli.setup import _info, print_info, print_success, print_warning, prompt_yes_no
if not _check_espeak_ng():
if not (shutil.which("espeak-ng") or shutil.which("espeak")):
hint, install_cmd = _ESPEAK_INSTALL.get(sys.platform, _ESPEAK_INSTALL["linux"])
print()
print_warning("NeuTTS requires espeak-ng for phonemization.")
@@ -72,10 +67,7 @@ def _install_neutts_deps() -> bool:
def _install_kittentts_deps() -> bool:
"""Install KittenTTS dependencies with user approval. Returns True on success."""
from hermes_cli.setup import _info
wheel_url = (
"https://github.com/KittenML/KittenTTS/releases/download/"
"0.8.1/kittentts-0.8.1-py3-none-any.whl"
)
wheel_url = "https://github.com/KittenML/KittenTTS/releases/download/0.8.1/kittentts-0.8.1-py3-none-any.whl"
_info(None, "Installing kittentts Python package (~25-80MB model downloaded on first use)...", None)
return _pip_install_tts_package(
"kittentts", ["-U", wheel_url, "soundfile", "--quiet"], f"uv pip install -U '{wheel_url}' soundfile",
@@ -127,11 +119,6 @@ def _run_xai_oauth_login_from_setup() -> bool:
return False
_TTS_PROVIDER_LABELS = {
"edge": "Edge TTS", "elevenlabs": "ElevenLabs", "openai": "OpenAI TTS", "xai": "xAI TTS",
"minimax": "MiniMax TTS", "mistral": "Mistral Voxtral TTS", "gemini": "Google Gemini TTS",
"neutts": "NeuTTS", "kittentts": "KittenTTS",
}
_TTS_PROVIDER_CHOICES = [
("edge", "Edge TTS (free, cloud-based, no setup needed)"),
("elevenlabs", "ElevenLabs (premium quality, needs API key)"),
@@ -143,42 +130,34 @@ _TTS_PROVIDER_CHOICES = [
("neutts", "NeuTTS (local on-device, free, ~300MB model download)"),
("kittentts", "KittenTTS (local on-device, free, lightweight ~25-80MB ONNX)"),
]
# Short label = menu label minus its parenthetical ("Edge TTS", "Mistral Voxtral TTS", ...).
_TTS_PROVIDER_LABELS = {key: label.split(" (")[0] for key, label in _TTS_PROVIDER_CHOICES}
# provider -> (env vars that satisfy it, env var to save, prompt, success line, pre-prompt hint)
_TTS_API_KEY_PROVIDERS = {
"elevenlabs": (("ELEVENLABS_API_KEY",), "ELEVENLABS_API_KEY", "ElevenLabs API key", "ElevenLabs API key saved", ""),
"openai": (
("VOICE_TOOLS_OPENAI_KEY", "OPENAI_API_KEY"), "VOICE_TOOLS_OPENAI_KEY",
"OpenAI API key for TTS", "OpenAI TTS API key saved", "",
),
"openai": (("VOICE_TOOLS_OPENAI_KEY", "OPENAI_API_KEY"), "VOICE_TOOLS_OPENAI_KEY",
"OpenAI API key for TTS", "OpenAI TTS API key saved", ""),
"minimax": (("MINIMAX_API_KEY",), "MINIMAX_API_KEY", "MiniMax API key for TTS", "MiniMax TTS API key saved", ""),
"mistral": (("MISTRAL_API_KEY",), "MISTRAL_API_KEY", "Mistral API key for TTS", "Mistral TTS API key saved", ""),
"gemini": (
("GEMINI_API_KEY", "GOOGLE_API_KEY"), "GEMINI_API_KEY", "Gemini API key for TTS", "Gemini TTS API key saved",
"Get a free API key at https://aistudio.google.com/app/apikey",
),
"gemini": (("GEMINI_API_KEY", "GOOGLE_API_KEY"), "GEMINI_API_KEY", "Gemini API key for TTS",
"Gemini TTS API key saved", "Get a free API key at https://aistudio.google.com/app/apikey"),
}
# provider -> (module, display name, requirement lines, install question, installer)
_TTS_LOCAL_PROVIDERS = {
"neutts": (
"neutts", "NeuTTS",
("NeuTTS requires:", " • Python package: neutts (~50MB install + ~300MB model on first use)",
" • System package: espeak-ng (phonemizer)"),
"Install NeuTTS dependencies now?", _install_neutts_deps,
),
"kittentts": (
"kittentts", "KittenTTS",
("KittenTTS is lightweight (~25-80MB, CPU-only, no API key required).",
"Voices: Jasper, Bella, Luna, Bruno, Rosie, Hugo, Kiki, Leo"),
"Install KittenTTS now?", _install_kittentts_deps,
),
"neutts": ("neutts", "NeuTTS",
("NeuTTS requires:", " • Python package: neutts (~50MB install + ~300MB model on first use)",
" • System package: espeak-ng (phonemizer)"),
"Install NeuTTS dependencies now?", _install_neutts_deps),
"kittentts": ("kittentts", "KittenTTS",
("KittenTTS is lightweight (~25-80MB, CPU-only, no API key required).",
"Voices: Jasper, Bella, Luna, Bruno, Rosie, Hugo, Kiki, Leo"),
"Install KittenTTS now?", _install_kittentts_deps),
}
def _tts_api_key_step(selected: str) -> str:
"""Ensure the key for an API-key TTS provider exists; fall back to edge otherwise."""
from hermes_cli.setup import (
get_env_value, print_info, print_success, print_warning, prompt, save_env_value,
)
from hermes_cli.setup import get_env_value, print_info, print_success, print_warning, prompt, save_env_value
env_vars, save_var, prompt_label, saved_msg, hint = _TTS_API_KEY_PROVIDERS[selected]
if any(get_env_value(v) for v in env_vars):
return selected
@@ -230,10 +209,8 @@ def _tts_xai_step(config: dict) -> str:
print()
choice_idx = prompt_choice(
"How do you want xAI TTS to authenticate?",
choices=[
"Sign in with xAI Grok OAuth (SuperGrok / Premium+) — browser login",
"Paste an xAI API key (console.x.ai)", "Skip → fallback to Edge TTS",
],
choices=["Sign in with xAI Grok OAuth (SuperGrok / Premium+) — browser login",
"Paste an xAI API key (console.x.ai)", "Skip → fallback to Edge TTS"],
default=0,
)
if choice_idx == 0:
@@ -249,11 +226,8 @@ def _tts_xai_step(config: dict) -> str:
print_success("xAI TTS API key saved")
else:
from hermes_constants import display_hermes_home as _dhh
print_warning(
"No xAI API key provided for TTS. Configure XAI_API_KEY "
f"via hermes setup model or {_dhh()}/.env to use xAI TTS. "
"Falling back to Edge TTS."
)
print_warning("No xAI API key provided for TTS. Configure XAI_API_KEY "
f"via hermes setup model or {_dhh()}/.env to use xAI TTS. Falling back to Edge TTS.")
selected = "edge"
else:
print_warning("xAI TTS skipped. Falling back to Edge TTS.")
@@ -261,10 +235,10 @@ def _tts_xai_step(config: dict) -> str:
if selected == "xai":
print()
voice_id = prompt("xAI voice_id (Enter for 'eve', or paste a custom voice ID)")
if voice_id and voice_id.strip():
config.setdefault("tts", {}).setdefault("xai", {})["voice_id"] = voice_id.strip()
print_success(f"xAI voice_id set to: {voice_id.strip()}")
voice_id = (prompt("xAI voice_id (Enter for 'eve', or paste a custom voice ID)") or "").strip()
if voice_id:
config.setdefault("tts", {}).setdefault("xai", {})["voice_id"] = voice_id
print_success(f"xAI voice_id set to: {voice_id}")
return selected
@@ -296,9 +270,8 @@ def _setup_tts_provider(config: dict):
selected = "openai"
print_info("OpenAI TTS will use the managed Nous gateway and bill to your subscription.")
if get_env_value("VOICE_TOOLS_OPENAI_KEY") or get_env_value("OPENAI_API_KEY"):
print_warning(
"Direct OpenAI credentials are still configured and may take precedence until removed from ~/.hermes/.env."
)
print_warning("Direct OpenAI credentials are still configured and may take precedence "
"until removed from ~/.hermes/.env.")
if selected in _TTS_LOCAL_PROVIDERS:
selected = _tts_local_install_step(selected)
@@ -315,5 +288,3 @@ def _setup_tts_provider(config: dict):
def setup_tts(config: dict):
"""Standalone TTS setup (for 'hermes setup tts')."""
_setup_tts_provider(config)