HKUDS/DeepTutor · error · VoiceProviderError

Unsupported STT adapter: {name!r}

Error message

Unsupported STT adapter: {name!r}

What it means

get_stt_adapter(name) resolves an STT adapter from the STT_ADAPTERS registry keyed by name, defaulting to "openai_compat" only when the name is empty/None. If the name is non-empty but not a registered key, a VoiceProviderError is raised. This guards against typos or uninstalled STT backends.

Source

Thrown at deeptutor/services/voice/adapters/__init__.py:38

    "openrouter_tts": OpenRouterTTSAdapter(),
}

STT_ADAPTERS: dict[str, BaseSTTAdapter] = {
    "openai_compat": OpenAICompatSTTAdapter(),
}


def get_tts_adapter(name: str) -> BaseTTSAdapter:
    adapter = TTS_ADAPTERS.get(name or "openai_compat")
    if adapter is None:
        raise VoiceProviderError(f"Unsupported TTS adapter: {name!r}")
    return adapter


def get_stt_adapter(name: str) -> BaseSTTAdapter:
    adapter = STT_ADAPTERS.get(name or "openai_compat")
    if adapter is None:
        raise VoiceProviderError(f"Unsupported STT adapter: {name!r}")
    return adapter


__all__ = [
    "TTS_ADAPTERS",
    "STT_ADAPTERS",
    "get_tts_adapter",
    "get_stt_adapter",
]

View on GitHub (pinned to 3e82f13042)

Solutions

1. Check the exact keys of STT_ADAPTERS in deeptutor/services/voice/adapters/__init__.py and correct the configured name
2. Set the STT adapter name to "openai_compat" (or leave it empty to use the default)
3. If a custom adapter is expected, register it in STT_ADAPTERS or install the extra that provides it

Example fix

// before
adapter = get_stt_adapter("whisper_local")
// after
adapter = get_stt_adapter("openai_compat")

from deeptutor.services.voice.adapters import STT_ADAPTERS

name = (stt_name or "openai_compat").strip()
if name not in STT_ADAPTERS:
    raise ValueError(f"Unknown STT adapter {name!r}; available: {sorted(STT_ADAPTERS)}")
adapter = get_stt_adapter(name)

def is_known_stt_adapter(name: str) -> bool:
    return (name or "openai_compat") in STT_ADAPTERS

When it happens

Trigger: Calling get_stt_adapter("whisper") or any name not present in STT_ADAPTERS (e.g. a name copied from TTS config, or a backend whose extra deps were not installed). Empty string or None falls back to "openai_compat" and does NOT raise.

Common situations: A voice settings JSON names an STT provider that isn't compiled into this build (missing optional dependency), a copy-paste of the TTS adapter name into the STT field, or a renamed adapter after an upgrade.

Related errors

• {action} failed with HTTP {status_code}: {detail}
• No endpoint URL configured for STT.
• No audio data to transcribe.
• STT request error: {exc}
• Transcription response had no `text` field.

AI-assisted analysis of HKUDS/DeepTutor@3e82f13042 (2026-08-27). Data as JSON: /api/errors/62b51b8812fd926a. Report an issue: GitHub.