Support scaling the duration of a pause in TTS. (#1820)

This commit is contained in:
Fangjun Kuang
2025-02-08 12:47:26 +08:00
committed by GitHub
parent d38cb81014
commit 69f489f0cd
24 changed files with 171 additions and 19 deletions

View File

@@ -1135,6 +1135,7 @@ static sherpa_onnx::OfflineTtsConfig GetOfflineTtsConfig(
tts_config.rule_fsts = SHERPA_ONNX_OR(config->rule_fsts, "");
tts_config.rule_fars = SHERPA_ONNX_OR(config->rule_fars, "");
tts_config.max_num_sentences = SHERPA_ONNX_OR(config->max_num_sentences, 1);
tts_config.silence_scale = SHERPA_ONNX_OR(config->silence_scale, 0.2);
if (tts_config.model.debug) {
#if __OHOS__

View File

@@ -944,6 +944,7 @@ SHERPA_ONNX_API typedef struct SherpaOnnxOfflineTtsConfig {
const char *rule_fsts;
int32_t max_num_sentences;
const char *rule_fars;
float silence_scale;
} SherpaOnnxOfflineTtsConfig;
SHERPA_ONNX_API typedef struct SherpaOnnxGeneratedAudio {

View File

@@ -352,6 +352,7 @@ OfflineTts OfflineTts::Create(const OfflineTtsConfig &config) {
c.rule_fsts = config.rule_fsts.c_str();
c.max_num_sentences = config.max_num_sentences;
c.silence_scale = config.silence_scale;
c.rule_fars = config.rule_fars.c_str();
auto p = SherpaOnnxCreateOfflineTts(&c);

View File

@@ -363,6 +363,7 @@ struct OfflineTtsConfig {
std::string rule_fsts;
std::string rule_fars;
int32_t max_num_sentences = 1;
float silence_scale = 0.2;
};
struct GeneratedAudio {