Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-31 14:33:54 -04:00
Code Issues Packages Projects Releases Wiki Activity

Make normalization configurable

Browse Source 
Still need to add an actual config UI for it. Will see if its actually
needed.
This commit is contained in:
Kovid Goyal 2025-07-30 17:27:02 +05:30
parent 6f9820e4d9
commit 5a03ee2e85
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
2 changed files with 3 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/calibre/utils/tts/piper.cpp
View File

@ -57,6 +57,7 @@ static float current_length_scale = 1;
static float current_noise_scale = 1; static float current_noise_scale = 1;
static float current_noise_w = 1; static float current_noise_w = 1;
static float current_sentence_delay = 0; static float current_sentence_delay = 0;
static bool current_normalize_volume = true;
std::unique_ptr<Ort::Session> session; std::unique_ptr<Ort::Session> session;
std::queue<std::vector<PhonemeId>> phoneme_id_queue; std::queue<std::vector<PhonemeId>> phoneme_id_queue;
std::vector<float> chunk_samples; std::vector<float> chunk_samples;
@ -231,6 +232,7 @@ set_voice(PyObject *self, PyObject *args) {
G(noise_scale, current_noise_scale, (float)PyFloat_AsDouble); G(noise_scale, current_noise_scale, (float)PyFloat_AsDouble);
G(noise_w, current_noise_w, (float)PyFloat_AsDouble); G(noise_w, current_noise_w, (float)PyFloat_AsDouble);
G(sentence_delay, current_sentence_delay, (float)PyFloat_AsDouble); G(sentence_delay, current_sentence_delay, (float)PyFloat_AsDouble);
G(normalize_volume, current_normalize_volume, PyObject_IsTrue);
#undef G #undef G
PyObject *map = PyObject_GetAttrString(cfg, "phoneme_id_map"); PyObject *map = PyObject_GetAttrString(cfg, "phoneme_id_map");

1
src/calibre/utils/tts/piper.py
View File

@ -28,6 +28,7 @@ class VoiceConfig(NamedTuple):
noise_w: float noise_w: float
num_speakers: int num_speakers: int
sentence_delay: float = 0 sentence_delay: float = 0
normalize_volume: bool = False
def translate_voice_config(x: Any) -> VoiceConfig: def translate_voice_config(x: Any) -> VoiceConfig: