langgenius/dify · warning · SpeechToTextDisabledError

speech_to_text_disabled

speech_to_text_disabled

Error message

Speech to text is disabled.

What it means

HTTP 400 SpeechToTextDisabledError, raised by the STT endpoint. The effective application configuration has speech-to-text turned off (or the config is missing entirely). AudioService.transcript_asr checks the feature per app mode: for ADVANCED_CHAT/WORKFLOW it requires workflow.features_dict['speech_to_text']['enabled']; for legacy modes it requires app_model_config.speech_to_text_dict['enabled']; for AGENT apps it merges the agent soul features and requires the merged speech_to_text enabled. Any of these being absent/false raises SpeechToTextDisabledServiceError.

Source

Thrown at api/controllers/console/explore/audio.py:81

                file=file,
                session=db.session(),
                end_user=None,
            )

            return response
        except services.errors.app_model_config.AppModelConfigBrokenError:
            logger.exception("App model config broken.")
            raise AppUnavailableError()
        except NoAudioUploadedServiceError:
            raise NoAudioUploadedError()
        except AudioTooLargeServiceError as e:
            raise AudioTooLargeError(str(e))
        except UnsupportedAudioTypeServiceError:
            raise UnsupportedAudioTypeError()
        except ProviderNotSupportSpeechToTextServiceError:
            raise ProviderNotSupportSpeechToTextError()
        except SpeechToTextDisabledServiceError:
            raise SpeechToTextDisabledError()
        except ProviderTokenNotInitError as ex:
            raise ProviderNotInitializeError(ex.description)
        except QuotaExceededError:
            raise ProviderQuotaExceededError()
        except ModelCurrentlyNotSupportError:
            raise ProviderModelCurrentlyNotSupportError()
        except InvokeError as e:
            raise CompletionRequestError(e.description)
        except ValueError as e:
            raise e
        except Exception as e:
            logger.exception("internal server error.")
            raise InternalServerError()


@console_ns.route(
    "/installed-apps/<uuid:installed_app_id>/text-to-audio",
    endpoint="installed_app_text",

View on GitHub (pinned to ef8544b173)

Solutions

1. Open the app in the studio, enable the Speech-to-Text feature, and republish (for workflow/advanced-chat) or save (for legacy chat/completion).
2. For Agent apps, ensure the published Agent Soul has speech_to_text enabled, then republish.
3. If the AppModelConfig row is missing, restore or recreate the app configuration.
4. Confirm you are calling audio-to-text on an app that is actually meant to support voice input.

// Only render the audio-upload control when the app's feature flags enable STT.
if (app.features?.speech_to_text?.enabled) renderMicButton()
else hideMicButton()

try { await postAudio(file) }
catch (e) { if (e.code === 400 && e.error_code === 'speech_to_text_disabled') disableMicButton() else throw e }

When it happens

Trigger: POST /console/explore/installed-apps/{id}/audio-to-text on an app whose STT feature flag is disabled; a workflow app with no 'speech_to_text' block in features_dict; a legacy app whose AppModelConfig row is missing or has speech_to_text_dict.enabled = false; a published Agent whose merged soul features disable STT.

Common situations: App author never enabled the Speech-to-Text toggle in the app studio; workflow was republished without the STT feature; app model config was corrupted/deleted; agent soul published without STT.

Related errors

• provider_not_support_speech_to_text
• provider_not_initialize
• provider_quota_exceeded
• model_currently_not_support
• completion_request_error

AI-assisted analysis of langgenius/dify@ef8544b173 (2026-08-12). Data as JSON: /api/errors/6138f7fb6d9b4647. Report an issue: GitHub.