@jambonz/schema 0.1.1

package/components/recognizer-elevenlabsOptions.schema.json

@@ -0,0 +1,39 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://jambonz.org/schema/components/recognizer-elevenlabsOptions",
+  "title": "ElevenLabs Recognizer Options",
+  "description": "ElevenLabs-specific STT options. Only applies when recognizer vendor is 'elevenlabs'.",
+  "type": "object",
+  "properties": {
+    "includeTimestamps": {
+      "type": "boolean",
+      "description": "Include word-level timestamps in results."
+    },
+    "commitStrategy": {
+      "type": "string",
+      "enum": ["manual", "vad"],
+      "description": "How audio chunks are committed. 'manual' for explicit commits, 'vad' for voice activity detection."
+    },
+    "vadSilenceThresholdSecs": {
+      "type": "number",
+      "description": "Silence duration in seconds to trigger VAD commit."
+    },
+    "vadThreshold": {
+      "type": "number",
+      "description": "VAD activation threshold."
+    },
+    "minSpeechDurationMs": {
+      "type": "number",
+      "description": "Minimum speech duration in milliseconds to accept."
+    },
+    "minSilenceDurationMs": {
+      "type": "number",
+      "description": "Minimum silence duration in milliseconds to trigger end of speech."
+    },
+    "enableLogging": {
+      "type": "boolean",
+      "description": "Enable server-side logging."
+    }
+  },
+  "additionalProperties": false
+}

package/components/recognizer-gladiaOptions.schema.json

@@ -0,0 +1,8 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://jambonz.org/schema/components/recognizer-gladiaOptions",
+  "title": "Gladia Recognizer Options",
+  "description": "Gladia-specific STT options. Only applies when recognizer vendor is 'gladia'. See Gladia API documentation for available options.",
+  "type": "object",
+  "additionalProperties": false
+}

package/components/recognizer-googleOptions.schema.json

@@ -0,0 +1,35 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://jambonz.org/schema/components/recognizer-googleOptions",
+  "title": "Google Recognizer Options",
+  "description": "Google Speech-to-Text specific options. Only applies when recognizer vendor is 'google'.",
+  "type": "object",
+  "properties": {
+    "serviceVersion": {
+      "type": "string",
+      "enum": ["v1", "v2"],
+      "description": "Google Speech-to-Text API version."
+    },
+    "recognizerId": {
+      "type": "string",
+      "description": "ID of a Google Speech recognizer resource (v2 only)."
+    },
+    "speechStartTimeoutMs": {
+      "type": "number",
+      "description": "Timeout in milliseconds to wait for speech to start."
+    },
+    "speechEndTimeoutMs": {
+      "type": "number",
+      "description": "Timeout in milliseconds to detect end of speech."
+    },
+    "enableVoiceActivityEvents": {
+      "type": "boolean",
+      "description": "Enable voice activity detection events."
+    },
+    "transcriptNormalization": {
+      "type": "array",
+      "description": "Array of transcript normalization rules."
+    }
+  },
+  "additionalProperties": false
+}

package/components/recognizer-houndifyOptions.schema.json

@@ -0,0 +1,53 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://jambonz.org/schema/components/recognizer-houndifyOptions",
+  "title": "Houndify Recognizer Options",
+  "description": "Houndify-specific STT options. Only applies when recognizer vendor is 'houndify'.",
+  "type": "object",
+  "properties": {
+    "requestInfo": {
+      "type": "object",
+      "description": "Houndify RequestInfo object with context data.",
+      "additionalProperties": true
+    },
+    "sampleRate": { "type": "number", "description": "Audio sample rate in Hz." },
+    "latitude": { "type": "number", "description": "User latitude for location-aware queries." },
+    "longitude": { "type": "number", "description": "User longitude for location-aware queries." },
+    "city": { "type": "string", "description": "User city." },
+    "state": { "type": "string", "description": "User state." },
+    "country": { "type": "string", "description": "User country." },
+    "timeZone": { "type": "string", "description": "User timezone." },
+    "domain": { "type": "string", "description": "Houndify domain." },
+    "audioEndpoint": { "type": "string", "description": "Custom audio endpoint URL." },
+    "maxSilenceSeconds": { "type": "number", "description": "Maximum silence before stopping." },
+    "maxSilenceAfterFullQuerySeconds": { "type": "number", "description": "Silence timeout after a complete query." },
+    "maxSilenceAfterPartialQuerySeconds": { "type": "number", "description": "Silence timeout after a partial query." },
+    "vadSensitivity": { "type": "number", "description": "VAD sensitivity level." },
+    "vadTimeout": { "type": "number", "description": "VAD timeout in milliseconds." },
+    "vadMode": { "type": "string", "description": "VAD mode." },
+    "vadVoiceMs": { "type": "number", "description": "Milliseconds of voice to trigger VAD." },
+    "vadSilenceMs": { "type": "number", "description": "Milliseconds of silence to trigger VAD." },
+    "vadDebug": { "type": "boolean", "description": "Enable VAD debug logging." },
+    "audioFormat": { "type": "string", "description": "Audio format." },
+    "enableNoiseReduction": { "type": "boolean", "description": "Enable noise reduction." },
+    "enableProfanityFilter": { "type": "boolean", "description": "Filter profanity." },
+    "enablePunctuation": { "type": "boolean", "description": "Enable punctuation." },
+    "enableCapitalization": { "type": "boolean", "description": "Enable capitalization." },
+    "confidenceThreshold": { "type": "number", "description": "Minimum confidence threshold." },
+    "enableDisfluencyFilter": { "type": "boolean", "description": "Filter disfluencies (um, uh)." },
+    "maxResults": { "type": "number", "description": "Maximum number of results." },
+    "enableWordTimestamps": { "type": "boolean", "description": "Include word timestamps." },
+    "maxAlternatives": { "type": "number", "description": "Maximum alternative transcripts." },
+    "partialTranscriptInterval": { "type": "number", "description": "Interval for partial transcript delivery." },
+    "sessionTimeout": { "type": "number", "description": "Session timeout." },
+    "connectionTimeout": { "type": "number", "description": "Connection timeout." },
+    "customVocabulary": {
+      "type": "array",
+      "items": { "type": "string" },
+      "description": "Custom vocabulary terms."
+    },
+    "languageModel": { "type": "string", "description": "Language model to use." },
+    "audioQueryAbsoluteTimeout": { "type": "number", "description": "Absolute timeout for audio queries." }
+  },
+  "additionalProperties": false
+}

package/components/recognizer-ibmOptions.schema.json

@@ -0,0 +1,54 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://jambonz.org/schema/components/recognizer-ibmOptions",
+  "title": "IBM Recognizer Options",
+  "description": "IBM Watson Speech-to-Text specific options. Only applies when recognizer vendor is 'ibm'.",
+  "type": "object",
+  "properties": {
+    "sttApiKey": {
+      "type": "string",
+      "description": "IBM STT API key. Overrides credentials configured in jambonz."
+    },
+    "sttRegion": {
+      "type": "string",
+      "description": "IBM STT region."
+    },
+    "ttsApiKey": {
+      "type": "string",
+      "description": "IBM TTS API key."
+    },
+    "ttsRegion": {
+      "type": "string",
+      "description": "IBM TTS region."
+    },
+    "instanceId": {
+      "type": "string",
+      "description": "IBM Watson instance ID."
+    },
+    "model": {
+      "type": "string",
+      "description": "Recognition model name."
+    },
+    "languageCustomizationId": {
+      "type": "string",
+      "description": "ID of a custom language model."
+    },
+    "acousticCustomizationId": {
+      "type": "string",
+      "description": "ID of a custom acoustic model."
+    },
+    "baseModelVersion": {
+      "type": "string",
+      "description": "Base model version to use."
+    },
+    "watsonMetadata": {
+      "type": "string",
+      "description": "Customer ID metadata for data labeling."
+    },
+    "watsonLearningOptOut": {
+      "type": "boolean",
+      "description": "Opt out of IBM data collection for service improvements."
+    }
+  },
+  "additionalProperties": false
+}

package/components/recognizer-nuanceOptions.schema.json

@@ -0,0 +1,150 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://jambonz.org/schema/components/recognizer-nuanceOptions",
+  "title": "Nuance Recognizer Options",
+  "description": "Nuance Mix specific options. Only applies when recognizer vendor is 'nuance'.",
+  "type": "object",
+  "properties": {
+    "clientId": {
+      "type": "string",
+      "description": "Nuance Mix client ID."
+    },
+    "secret": {
+      "type": "string",
+      "description": "Nuance Mix client secret."
+    },
+    "kryptonEndpoint": {
+      "type": "string",
+      "description": "Custom Nuance Krypton endpoint URL."
+    },
+    "topic": {
+      "type": "string",
+      "description": "Recognition topic (domain)."
+    },
+    "utteranceDetectionMode": {
+      "type": "string",
+      "enum": ["single", "multiple", "disabled"],
+      "description": "How utterance boundaries are detected."
+    },
+    "punctuation": {
+      "type": "boolean",
+      "description": "Enable automatic punctuation."
+    },
+    "profanityFilter": {
+      "type": "boolean",
+      "description": "Filter profanity from results."
+    },
+    "includeTokenization": {
+      "type": "boolean",
+      "description": "Include tokenization data in results."
+    },
+    "discardSpeakerAdaptation": {
+      "type": "boolean",
+      "description": "Discard speaker adaptation data."
+    },
+    "suppressCallRecording": {
+      "type": "boolean",
+      "description": "Suppress call recording on the Nuance side."
+    },
+    "maskLoadFailures": {
+      "type": "boolean",
+      "description": "Mask resource load failures."
+    },
+    "suppressInitialCapitalization": {
+      "type": "boolean",
+      "description": "Suppress initial capitalization of results."
+    },
+    "allowZeroBaseLmWeight": {
+      "type": "boolean",
+      "description": "Allow zero base language model weight."
+    },
+    "filterWakeupWord": {
+      "type": "boolean",
+      "description": "Filter wakeup words from results."
+    },
+    "resultType": {
+      "type": "string",
+      "enum": ["final", "partial", "immutable_partial"],
+      "description": "Type of results to return."
+    },
+    "noInputTimeoutMs": {
+      "type": "number",
+      "description": "Timeout in milliseconds before no-input event."
+    },
+    "recognitionTimeoutMs": {
+      "type": "number",
+      "description": "Maximum recognition duration in milliseconds."
+    },
+    "utteranceEndSilenceMs": {
+      "type": "number",
+      "description": "Silence duration in milliseconds to detect end of utterance."
+    },
+    "maxHypotheses": {
+      "type": "number",
+      "description": "Maximum number of recognition hypotheses to return."
+    },
+    "speechDomain": {
+      "type": "string",
+      "description": "Speech domain for optimized recognition."
+    },
+    "formatting": {
+      "type": "object",
+      "description": "Formatting options for recognition results.",
+      "properties": {
+        "scheme": { "type": "string", "description": "Formatting scheme name." },
+        "options": { "type": "object", "description": "Scheme-specific formatting options." }
+      },
+      "required": ["scheme", "options"]
+    },
+    "clientData": {
+      "type": "object",
+      "description": "Custom client data to pass to Nuance.",
+      "additionalProperties": true
+    },
+    "userId": {
+      "type": "string",
+      "description": "User ID for speaker adaptation."
+    },
+    "speechDetectionSensitivity": {
+      "type": "number",
+      "description": "Speech detection sensitivity (0-1)."
+    },
+    "resources": {
+      "type": "array",
+      "description": "Array of Nuance recognition resources (grammars, wordsets, etc.).",
+      "items": {
+        "type": "object",
+        "properties": {
+          "externalReference": {
+            "type": "object",
+            "description": "External resource reference.",
+            "properties": {
+              "type": {
+                "type": "string",
+                "enum": ["undefined_resource_type", "wordset", "compiled_wordset", "domain_lm", "speaker_profile", "grammar", "settings"]
+              },
+              "uri": { "type": "string" },
+              "maxLoadFailures": { "type": "boolean" },
+              "requestTimeoutMs": { "type": "number" },
+              "headers": { "type": "object" }
+            }
+          },
+          "inlineWordset": { "type": "string", "description": "Inline wordset JSON string." },
+          "builtin": { "type": "string", "description": "Built-in grammar name." },
+          "inlineGrammar": { "type": "string", "description": "Inline SRGS grammar." },
+          "wakeupWord": { "type": "array", "items": { "type": "string" }, "description": "Wakeup words." },
+          "weightName": {
+            "type": "string",
+            "enum": ["defaultWeight", "lowest", "low", "medium", "high", "highest"]
+          },
+          "weightValue": { "type": "number" },
+          "reuse": {
+            "type": "string",
+            "enum": ["undefined_reuse", "low_reuse", "high_reuse"]
+          }
+        }
+      }
+    }
+  },
+  "additionalProperties": false
+}

package/components/recognizer-nvidiaOptions.schema.json

@@ -0,0 +1,39 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://jambonz.org/schema/components/recognizer-nvidiaOptions",
+  "title": "NVIDIA Recognizer Options",
+  "description": "NVIDIA Riva specific options. Only applies when recognizer vendor is 'nvidia'.",
+  "type": "object",
+  "properties": {
+    "rivaUri": {
+      "type": "string",
+      "description": "NVIDIA Riva server URI."
+    },
+    "maxAlternatives": {
+      "type": "number",
+      "description": "Maximum number of alternative transcripts."
+    },
+    "profanityFilter": {
+      "type": "boolean",
+      "description": "Filter profanity from results."
+    },
+    "punctuation": {
+      "type": "boolean",
+      "description": "Enable automatic punctuation."
+    },
+    "wordTimeOffsets": {
+      "type": "boolean",
+      "description": "Include word-level timestamps."
+    },
+    "verbatimTranscripts": {
+      "type": "boolean",
+      "description": "Return verbatim (unformatted) transcripts."
+    },
+    "customConfiguration": {
+      "type": "object",
+      "description": "Custom Riva configuration parameters.",
+      "additionalProperties": true
+    }
+  },
+  "additionalProperties": false
+}

package/components/recognizer-openaiOptions.schema.json

@@ -0,0 +1,59 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://jambonz.org/schema/components/recognizer-openaiOptions",
+  "title": "OpenAI Recognizer Options",
+  "description": "OpenAI Whisper/Realtime specific STT options. Only applies when recognizer vendor is 'openai'.",
+  "type": "object",
+  "properties": {
+    "apiKey": {
+      "type": "string",
+      "description": "OpenAI API key. Overrides credentials configured in jambonz."
+    },
+    "model": {
+      "type": "string",
+      "description": "OpenAI STT model name."
+    },
+    "prompt": {
+      "type": "string",
+      "description": "Prompt to guide the recognition model."
+    },
+    "promptTemplates": {
+      "type": "object",
+      "description": "Templates for dynamic prompt generation.",
+      "properties": {
+        "hintsTemplate": { "type": "string", "description": "Template for injecting hints into the prompt." },
+        "conversationHistoryTemplate": { "type": "string", "description": "Template for injecting conversation history." }
+      }
+    },
+    "language": {
+      "type": "string",
+      "description": "Language code for recognition."
+    },
+    "input_audio_noise_reduction": {
+      "type": "string",
+      "enum": ["near_field", "far_field"],
+      "description": "Input audio noise reduction mode."
+    },
+    "turn_detection": {
+      "type": "object",
+      "description": "Turn detection configuration for the OpenAI Realtime API.",
+      "properties": {
+        "type": {
+          "type": "string",
+          "enum": ["none", "server_vad", "semantic_vad"],
+          "description": "Turn detection strategy."
+        },
+        "eagerness": {
+          "type": "string",
+          "enum": ["low", "medium", "high", "auto"],
+          "description": "How eagerly the model should respond."
+        },
+        "threshold": { "type": "number", "description": "VAD activation threshold (0-1)." },
+        "prefix_padding_ms": { "type": "number", "description": "Milliseconds of audio to include before detected speech." },
+        "silence_duration_ms": { "type": "number", "description": "Milliseconds of silence to detect end of speech." }
+      },
+      "required": ["type"]
+    }
+  },
+  "additionalProperties": false
+}

package/components/recognizer-sonioxOptions.schema.json

@@ -0,0 +1,46 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://jambonz.org/schema/components/recognizer-sonioxOptions",
+  "title": "Soniox Recognizer Options",
+  "description": "Soniox-specific STT options. Only applies when recognizer vendor is 'soniox'.",
+  "type": "object",
+  "properties": {
+    "apiKey": {
+      "type": "string",
+      "description": "Soniox API key."
+    },
+    "model": {
+      "type": "string",
+      "description": "Soniox recognition model."
+    },
+    "endpointDetection": {
+      "type": "boolean",
+      "description": "Enable endpoint detection."
+    },
+    "profanityFilter": {
+      "type": "boolean",
+      "description": "Filter profanity from results."
+    },
+    "speechContext": {
+      "type": "string",
+      "description": "Speech context for improved recognition."
+    },
+    "clientRequestReference": {
+      "type": "string",
+      "description": "Client request reference for tracking."
+    },
+    "storage": {
+      "type": "object",
+      "description": "Soniox storage configuration for persisting transcripts.",
+      "properties": {
+        "id": { "type": "string", "description": "Storage ID." },
+        "title": { "type": "string", "description": "Storage title." },
+        "disableStoreAudio": { "type": "boolean", "description": "Disable audio storage." },
+        "disableStoreTranscript": { "type": "boolean", "description": "Disable transcript storage." },
+        "disableSearch": { "type": "boolean", "description": "Disable search indexing." },
+        "metadata": { "type": "object", "description": "Custom metadata.", "additionalProperties": true }
+      }
+    }
+  },
+  "additionalProperties": false
+}

package/components/recognizer-speechmaticsOptions.schema.json

@@ -0,0 +1,100 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://jambonz.org/schema/components/recognizer-speechmaticsOptions",
+  "title": "Speechmatics Recognizer Options",
+  "description": "Speechmatics-specific STT options. Only applies when recognizer vendor is 'speechmatics'.",
+  "type": "object",
+  "properties": {
+    "host": {
+      "type": "string",
+      "description": "Speechmatics host URL."
+    },
+    "profile": {
+      "type": "string",
+      "enum": ["adaptive", "agile", "smart", "external"],
+      "description": "Speechmatics profile."
+    },
+    "transcription_config": {
+      "type": "object",
+      "description": "Speechmatics transcription configuration.",
+      "properties": {
+        "language": { "type": "string", "description": "Language code." },
+        "domain": { "type": "string", "description": "Domain model." },
+        "additional_vocab": { "type": "array", "description": "Additional vocabulary entries." },
+        "diarization": { "type": "string", "description": "Diarization mode." },
+        "speaker_diarization_config": {
+          "type": "object",
+          "properties": {
+            "speaker_sensitivity": { "type": "number" },
+            "max_speakers": { "type": "number" }
+          }
+        },
+        "conversation_config": {
+          "type": "object",
+          "properties": {
+            "end_of_utterance_silence_trigger": { "type": "number" }
+          }
+        },
+        "enable_partials": { "type": "boolean", "description": "Enable partial transcripts." },
+        "max_delay": { "type": "number", "description": "Maximum delay in seconds." },
+        "max_delay_mode": {
+          "type": "string",
+          "enum": ["fixed", "flexible"],
+          "description": "Delay mode."
+        },
+        "output_locale": { "type": "string", "description": "Output locale for formatting." },
+        "punctuation_overrides": {
+          "type": "object",
+          "properties": {
+            "permitted_marks": { "type": "array", "items": { "type": "string" } },
+            "sensitivity": { "type": "number" }
+          }
+        },
+        "operating_point": { "type": "string", "description": "Operating point (standard or enhanced)." },
+        "enable_entities": { "type": "boolean", "description": "Enable entity detection." },
+        "audio_filtering_config": {
+          "type": "object",
+          "properties": {
+            "volume_threshold": { "type": "number" }
+          },
+          "required": ["volume_threshold"]
+        },
+        "transcript_filtering_config": {
+          "type": "object",
+          "properties": {
+            "remove_disfluencies": { "type": "boolean" }
+          },
+          "required": ["remove_disfluencies"]
+        }
+      }
+    },
+    "translation_config": {
+      "type": "object",
+      "description": "Speechmatics translation configuration.",
+      "properties": {
+        "target_languages": {
+          "type": "array",
+          "items": { "type": "string" },
+          "description": "Target languages for translation."
+        },
+        "enable_partials": { "type": "boolean", "description": "Enable partial translations." }
+      },
+      "required": ["target_languages"]
+    },
+    "audio_events_config": {
+      "type": "object",
+      "description": "Audio event detection configuration.",
+      "properties": {
+        "types": {
+          "type": "array",
+          "items": {
+            "type": "string",
+            "enum": ["applause", "music", "laughter"]
+          },
+          "description": "Audio event types to detect."
+        }
+      }
+    }
+  },
+  "additionalProperties": false
+}

package/components/recognizer-verbioOptions.schema.json

@@ -0,0 +1,46 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://jambonz.org/schema/components/recognizer-verbioOptions",
+  "title": "Verbio Recognizer Options",
+  "description": "Verbio-specific STT options. Only applies when recognizer vendor is 'verbio'.",
+  "type": "object",
+  "properties": {
+    "enable_formatting": {
+      "type": "boolean",
+      "description": "Enable text formatting of results."
+    },
+    "enable_diarization": {
+      "type": "boolean",
+      "description": "Enable speaker diarization."
+    },
+    "topic": {
+      "type": "number",
+      "description": "Topic ID for domain-specific recognition."
+    },
+    "inline_grammar": {
+      "type": "string",
+      "description": "Inline SRGS grammar for constrained recognition."
+    },
+    "grammar_uri": {
+      "type": "string",
+      "description": "URI of an external grammar resource."
+    },
+    "label": {
+      "type": "string",
+      "description": "Label for the recognition session."
+    },
+    "recognition_timeout": {
+      "type": "number",
+      "description": "Maximum recognition duration in seconds."
+    },
+    "speech_complete_timeout": {
+      "type": "number",
+      "description": "Silence duration in seconds after complete speech."
+    },
+    "speech_incomplete_timeout": {
+      "type": "number",
+      "description": "Silence duration in seconds after incomplete speech."
+    }
+  },
+  "additionalProperties": false
+}