Interface: SynthesizeSpeechRequest

@kortexya/reasoninglayer

---

@kortexya/reasoninglayer / SynthesizeSpeechRequest

Interface: SynthesizeSpeechRequest

Defined in: src/types/speech.ts:117

Body of POST /api/v1/speech/synthesize — synthesize speech.

Properties

description?

optional description?: string | null

Defined in: src/types/speech.ts:121

VoiceDesign: describe the desired voice in words. Optional.

---

engine?

optional engine?: string | null

Defined in: src/types/speech.ts:123

Engine to render with ("qwen3-tts", "chatterbox"). Defaults to "qwen3-tts". Optional.

---

format?

optional format?: string | null

Defined in: src/types/speech.ts:125

Output encoding: "wav" (default), "pcm_s16le", "pcm_f32le". Optional.

---

instruction?

optional instruction?: string | null

Defined in: src/types/speech.ts:127

Style instruction accompanying speaker. Optional.

---

language?

optional language?: string | null

Defined in: src/types/speech.ts:129

Language hint. Optional.

---

maxDurationS?

optional maxDurationS?: number | null

Defined in: src/types/speech.ts:131

Cap on generated audio length in seconds. Optional.

---

refAudioBase64?

optional refAudioBase64?: string | null

Defined in: src/types/speech.ts:119

Ad-hoc zero-shot clone from a base64 reference clip. Optional.

---

refText?

optional refText?: string | null

Defined in: src/types/speech.ts:133

Transcript of refAudioBase64, enabling the higher-quality ICL clone path. Optional.

---

seed?

optional seed?: number | null

Defined in: src/types/speech.ts:135

Sampling seed; fixing it makes synthesis reproducible. Optional.

---

speaker?

optional speaker?: string | null

Defined in: src/types/speech.ts:137

A checkpoint preset speaker name (see listEngines). Optional.

---

text

text: string

Defined in: src/types/speech.ts:139

Text to speak.

---

voiceId?

optional voiceId?: string | null

Defined in: src/types/speech.ts:141

A voice enrolled through enrollVoice, by its OSFKB id. Optional.