HKUDS/DeepTutor · error · VoiceProviderError

No endpoint URL configured for STT.

Error message

No endpoint URL configured for STT.

What it means

OpenAICompatSTTAdapter.transcribe requires config.base_url before it can build the {base}/audio/transcriptions URL; an empty/None base_url raises VoiceProviderError immediately. Configuration guard for the speech-to-text path.

Source

Thrown at deeptutor/services/voice/adapters/openai_compat.py:304

class OpenAICompatSTTAdapter(BaseSTTAdapter):
    """POST ``{base}/audio/transcriptions``.

    Multipart ``file`` upload by default; OpenRouter uses a base64-JSON body
    (``request_style == "base64_json"``) sharing the same path.
    """

    async def transcribe(
        self,
        audio: bytes,
        config: STTConfig,
        *,
        filename: str = "audio.webm",
        content_type: str = "application/octet-stream",
    ) -> str:
        if not config.base_url:
            raise VoiceProviderError("No endpoint URL configured for STT.")
        if not audio:
            raise VoiceProviderError("No audio data to transcribe.")
        url = join_audio_path(config.base_url, "audio/transcriptions")
        auth = build_auth_headers(config.auth_style, config.api_key)

        try:
            async with httpx.AsyncClient(timeout=config.request_timeout) as client:
                if config.request_style == STT_BASE64_JSON:
                    resp = await self._post_base64(client, url, auth, audio, filename, config)
                else:
                    resp = await self._post_multipart(
                        client, url, auth, audio, filename, content_type, config
                    )
        except httpx.HTTPError as exc:
            raise VoiceProviderError(f"STT request error: {exc}") from exc
        _raise_for_provider(resp, "Transcription")
        return self._parse_text(resp)

View on GitHub (pinned to 3e82f13042)

Solutions

1. Set base_url on the STTConfig / voice settings (e.g. https://api.openai.com/v1)
2. Check data/user/settings/*.json and env overrides for an empty STT endpoint
3. Validate voice provider config at startup

Example fix

// before
cfg = STTConfig(api_key=key)
// after
cfg = STTConfig(base_url="https://api.openai.com/v1", api_key=key)

if not (stt_config.base_url or "").strip():
    raise ValueError("STT base_url must be configured")

try:
    text = await stt.transcribe(audio, stt_config)
except VoiceProviderError as exc:
    if "No endpoint URL configured for STT" in str(exc):
        alert_admin_missing_stt_endpoint()  # config error, no retry

When it happens

Trigger: Calling transcribe(audio, config) with STTConfig.base_url empty — e.g. STT provider settings only contain an API key, or the env override cleared the endpoint.

Common situations: Assuming a default OpenAI endpoint exists without configuring it, settings JSON missing the stt.base_url field after an upgrade, or copy-pasting a key-only config from another tool.

Related errors

• No endpoint URL configured for TTS.
• No endpoint URL configured for this provider.
• Unsupported STT adapter: {name!r}
• No endpoint URL configured for this provider.
• {action} failed with HTTP {status_code}: {detail}

AI-assisted analysis of HKUDS/DeepTutor@3e82f13042 (2026-08-27). Data as JSON: /api/errors/e592b5c55c63c53e. Report an issue: GitHub.