Fabio
c36510c21e
Some checks failed
Linux Build / Linux cpu (push) Has been cancelled
Linux Build / Linux cpu-clang (push) Has been cancelled
Linux Build / Linux vulkan (push) Has been cancelled
mac-build / macOS CPU (push) Has been cancelled
Nix Build / Nix (cpu) (push) Has been cancelled
Nix Build / Nix (vulkan) (push) Has been cancelled
windows-build / Windows CPU (push) Has been cancelled
21 lines
699 B
C++
21 lines
699 B
C++
#pragma once
|
|
|
|
#include "engine/framework/runtime/session.h"
|
|
|
|
namespace minitts::server {
|
|
|
|
// On /v1/audio/speech/live, `input` is the text a text-to-speech model speaks,
|
|
// so those models need it. A model that answers or converts the live audio
|
|
// (s2s, vc, svc) takes it as an optional prompt and has its own default.
|
|
inline bool live_speech_requires_input(engine::runtime::VoiceTaskKind task) {
|
|
switch (task) {
|
|
case engine::runtime::VoiceTaskKind::SpeechToSpeech:
|
|
case engine::runtime::VoiceTaskKind::VoiceConversion:
|
|
case engine::runtime::VoiceTaskKind::Svc:
|
|
return false;
|
|
default:
|
|
return true;
|
|
}
|
|
}
|
|
|
|
} // namespace minitts::server
|