Class UTryllGenerateAndSpeakParams¶
ClassList > UTryllGenerateAndSpeakParams
#include <TryllGenerateAndSpeakParams.h>
Inherits the following classes: UTryllNodeParamsBase
Public Attributes¶
| Type | Name |
|---|---|
| int32 | ContextSize = 0 |
| FString | DefaultExit |
| FString | Grammar |
| ETryllHistoryRole | HistoryRole = ETryllHistoryRole::Assistant |
| FTryllSlotInput | Input |
| int32 | MinSentenceChars = 12 |
| FTryllModelName | ModelName |
| FTryllOutputFilterOverrides | OutputFilter |
| FTryllSlotName | OutputName |
| ETryllPlacement | Placement = ETryllPlacement::BeforeUserAsSystem |
| FTryllSamplingOverrides | Sampling |
| int32 | SpeakerId = 0 |
| float | Speed = 1.0f |
| FString | SystemPrompt |
| FString | Template |
| FString | TtsLang |
| FTryllModelName | TtsModelName |
| FTryllStoragePath | TtsVoice |
| bool | bOverrideGrammar = false |
| bool | bOverrideInput = false |
| bool | bOverrideModelName = false |
| bool | bOverrideOutputName = false |
| bool | bOverrideSystemPrompt = false |
| bool | bOverrideTemplate = false |
| bool | bOverrideTtsLang = false |
| bool | bOverrideTtsModelName = false |
| bool | bSubstituteAgentVariables = false |
Public Functions¶
| Type | Name |
|---|---|
| virtual ETryllNodeType | GetNodeType () override const |
| virtual flatbuffers::Offset< void > | Pack (flatbuffers::FlatBufferBuilder & Fbb, Tryll::NodeParams::NodeParams & OutType) override const |
Public Functions inherited from UTryllNodeParamsBase¶
| Type | Name |
|---|---|
| virtual TArray< FString > | GetExitTargetOptions () const |
| virtual ETryllNodeType | GetNodeType () const |
| FString | GetOwnerNodeName () const |
| virtual flatbuffers::Offset< void > | Pack (flatbuffers::FlatBufferBuilder & Fbb, ::Tryll::NodeParams::NodeParams & OutType) const |
| ETryllNodeType return | static_cast (0) |
Public Static Functions inherited from UTryllNodeParamsBase¶
| Type | Name |
|---|---|
| const FTryllGraphDescription * | FindEnclosingGraph (const UObject * Obj) |
Detailed Description¶
LLM-driven text generation + streaming TTS synthesis in a single fused node. Streams LLM tokens, segments them into sentences, synthesizes each sentence on a dedicated TTS strand, and emits PCM frames via OnTtsAudio callbacks.
Public Attributes Documentation¶
variable ContextSize¶
KV-cache / context window (n_ctx) for the LLM in tokens. 0 = fall back to the model variant's context_size, else the server default_n_ctx. Validated against the model's trained maximum at agent creation.
variable DefaultExit¶
Default exit target — routes here after generation + synthesis complete. Empty string = END. Graph exit "default" — target node name; empty = END.
variable Grammar¶
variable HistoryRole¶
Controls whether/how this node's slot replays in later turns' projected transcript. No send field — GenerateAndSpeak always streams text (streaming and non-streaming delivery would need incompatible projection-replay handling, so only the streamed mode is supported).
variable Input¶
Upstream inter-node slot this field consumes. Empty = context default.
variable MinSentenceChars¶
Minimum characters before a sentence boundary fires. Lower values reduce first-utterance latency at the cost of more, smaller TTS calls.
variable ModelName¶
variable OutputFilter¶
Opt-in bounded artifact cleanup applied to streamed LLM output before variable substitution, slot/history write, wire delivery, and TTS. Missing/null = all flags off (defaults). Appended after substitute_agent_variables for additive field ids.
variable OutputName¶
Effective output slot name for this node. Empty = the node's own name.
variable Placement¶
Where the rendered template body is placed relative to the resolved input.
variable Sampling¶
Sparse sampling overrides applied on top of the model-catalog defaults.
variable SpeakerId¶
TTS speaker / voice index (default 0). Valid range is 0 to the model's speaker count minus 1 — the shipped Supertonic 3 bundle has 10 voice styles (0–9); Pocket TTS is single-speaker, so it must stay 0 there (pick a Pocket voice with tts_voice instead). Agent creation fails with an out-of-range error for values the model cannot satisfy.
variable Speed¶
TTS speech-rate multiplier. 1.0 = native speed.
variable SystemPrompt¶
variable Template¶
variable TtsLang¶
variable TtsModelName¶
variable TtsVoice¶
Reference-voice WAV for zero-shot voice-cloning families (Pocket TTS): a path relative to the session storage root pointing at a 10–20 s clean mono recording of the target voice. Empty = the model's default voice. Ignored by fixed-voice families (use speaker_id there).
variable bOverrideGrammar¶
Optional GBNF grammar. When non-empty, the generated (and spoken) output is constrained to this grammar at every decode step. Empty = unconstrained. Must contain a root rule. Mutable — rebuilt per turn. Narrower fit than on Generate: use it to constrain spoken output to a fixed set of barks / canned lines. Validated at agent creation and on mutation (3007). Set to true to override the inherited Grammar value.
variable bOverrideInput¶
Slot name this node consumes as its primary text. Empty = "user_message". Structural: immutable after creation — rebinding would re-wire the slot dataflow that is validated once at agent creation. Set to true to override the inherited Input value.
variable bOverrideModelName¶
LLM model catalog name. Empty = use the agent's default_model_name. Set to true to override the inherited ModelName value.
variable bOverrideOutputName¶
Name this node's output slot is stored under. Empty = the node's own name. Structural: immutable after creation — renaming would re-wire the slot dataflow that is validated once at agent creation. Set to true to override the inherited OutputName value.
variable bOverrideSystemPrompt¶
Prepended before the user turn during projection. Set to true to override the inherited SystemPrompt value.
variable bOverrideTemplate¶
Mustache template applied to the user-area message at projection time. Set to true to override the inherited Template value.
variable bOverrideTtsLang¶
TTS language hint for multilingual models that select the language per call (e.g. Supertonic 3: "en", "de", "fr", …). Empty = the model's catalog default. Ignored by families whose language is fixed by the model itself (e.g. Pocket TTS per-language models). Set to true to override the inherited TtsLang value.
variable bOverrideTtsModelName¶
TTS model catalog name. Optional — when empty the server uses the first available TTS voice in the model catalog. Set it to pick a specific voice. Set to true to override the inherited TtsModelName value.
variable bSubstituteAgentVariables¶
When true, replace __VARIABLE__ markers in streamed LLM output with the rendered values of agent variables that declared allow_output_substitution. Matching is ASCII case-insensitive. Default false. Transformed text is authoritative for the slot, wire answer, history, and TTS input (substitution runs before synthesis). Appended after default_exit to keep FlatBuffers field ids additive.
Public Functions Documentation¶
function GetNodeType¶
Implements UTryllNodeParamsBase::GetNodeType
function Pack¶
virtual flatbuffers::Offset< void > UTryllGenerateAndSpeakParams::Pack (
flatbuffers::FlatBufferBuilder & Fbb,
Tryll::NodeParams::NodeParams & OutType
) override const
Implements UTryllNodeParamsBase::Pack
The documentation for this class was generated from the following file C:/_tryll/_monorepo3/tryll/clients/unreal/Source/TryllClient/Public/Generated/Nodes/TryllGenerateAndSpeakParams.h