mirror of
https://github.com/rhasspy/espeak-ng.git
synced 2026-10-31 05:35:35 +00:00
SpeechSynthesis.java: expose the different voice options from eSpeak
This makes the following changes:
1. setVoice takes a Voice object instead of a language string.
This is to make the API cleaner and allows SpeechSynthesis
to pass the correct parameter (identifier for SetVoiceByName,
or name for SetVoiceByProperties) to espeak.
2. The espeak_SetVoiceByName API is also exposed to support passing
the voice variant to use.
3. The age parameter has been re-added with the constants:
* SpeechSynthesis.AGE_ANY
* SpeechSynthesis.AGE_YOUNG
* SpeechSynthesis.AGE_OLD
based on the behaviour of the eSpeak voice selection algorithm.
NOTE: Due to the way that voice selection is implemented in eSpeak,
if variant is specified, the age and gender cannot be specified and
vice versa.
This commit is contained in:
@@ -241,16 +241,38 @@ JNICALL Java_com_reecedunn_espeak_SpeechSynthesis_nativeGetAvailableVoices(
|
||||
}
|
||||
|
||||
JNIEXPORT jboolean
|
||||
JNICALL Java_com_reecedunn_espeak_SpeechSynthesis_nativeSetVoice(
|
||||
JNIEnv *env, jobject object, jstring language, jint gender) {
|
||||
JNICALL Java_com_reecedunn_espeak_SpeechSynthesis_nativeSetVoiceByName(
|
||||
JNIEnv *env, jobject object, jstring name) {
|
||||
const char *c_name = name ? env->GetStringUTFChars(name, NULL) : NULL;
|
||||
|
||||
if (DEBUG) LOGV("%s(name=%s)", __FUNCTION__, c_name);
|
||||
|
||||
const espeak_ERROR result = espeak_SetVoiceByName(c_name);
|
||||
|
||||
if (c_name) env->ReleaseStringUTFChars(name, c_name);
|
||||
|
||||
switch (result) {
|
||||
case EE_OK: return JNI_TRUE;
|
||||
case EE_INTERNAL_ERROR: LOGE("espeak_SetVoiceByName: internal error."); break;
|
||||
case EE_BUFFER_FULL: LOGE("espeak_SetVoiceByName: buffer full."); break;
|
||||
case EE_NOT_FOUND: LOGE("espeak_SetVoiceByName: not found."); break;
|
||||
}
|
||||
|
||||
return JNI_FALSE;
|
||||
}
|
||||
|
||||
JNIEXPORT jboolean
|
||||
JNICALL Java_com_reecedunn_espeak_SpeechSynthesis_nativeSetVoiceByProperties(
|
||||
JNIEnv *env, jobject object, jstring language, jint gender, jint age) {
|
||||
const char *c_language = language ? env->GetStringUTFChars(language, NULL) : NULL;
|
||||
|
||||
if (DEBUG) LOGV("%s(language=%s, gender=%d)", __FUNCTION__, c_language, gender);
|
||||
if (DEBUG) LOGV("%s(language=%s, gender=%d, age=%d)", __FUNCTION__, c_language, gender, age);
|
||||
|
||||
espeak_VOICE voice_select;
|
||||
memset(&voice_select, 0, sizeof(espeak_VOICE));
|
||||
voice_select.languages = c_language;
|
||||
voice_select.gender = (int) gender;
|
||||
voice_select.age = (int) age;
|
||||
|
||||
const espeak_ERROR result = espeak_SetVoiceByProperties(&voice_select);
|
||||
|
||||
|
||||
@@ -45,6 +45,10 @@ public class SpeechSynthesis {
|
||||
public static final int GENDER_MALE = 1;
|
||||
public static final int GENDER_FEMALE = 2;
|
||||
|
||||
public static final int AGE_ANY = 0;
|
||||
public static final int AGE_YOUNG = 12;
|
||||
public static final int AGE_OLD = 60;
|
||||
|
||||
static {
|
||||
System.loadLibrary("ttsespeak");
|
||||
|
||||
@@ -169,8 +173,14 @@ public class SpeechSynthesis {
|
||||
return voices;
|
||||
}
|
||||
|
||||
public void setVoice(String language, int gender) {
|
||||
nativeSetVoice(language, gender);
|
||||
public void setVoice(Voice voice, String variant, int gender, int age) {
|
||||
// NOTE: espeak_SetVoiceByProperties does not support specifying the
|
||||
// voice variant (e.g. klatt), but espeak_SetVoiceByName does.
|
||||
if (variant == null) {
|
||||
nativeSetVoiceByProperties(voice.name, gender, age);
|
||||
} else {
|
||||
nativeSetVoiceByName(voice.identifier + "+" + variant);
|
||||
}
|
||||
}
|
||||
|
||||
public void setRate(int rate) {
|
||||
@@ -300,7 +310,9 @@ public class SpeechSynthesis {
|
||||
|
||||
private native final String[] nativeGetAvailableVoices();
|
||||
|
||||
private native final boolean nativeSetVoice(String language, int gender);
|
||||
private native final boolean nativeSetVoiceByName(String name);
|
||||
|
||||
private native final boolean nativeSetVoiceByProperties(String language, int gender, int age);
|
||||
|
||||
private native final boolean nativeSetRate(int rate);
|
||||
|
||||
|
||||
@@ -221,7 +221,7 @@ public class TtsService extends TextToSpeechService {
|
||||
mCallback.start(mEngine.getSampleRate(), mEngine.getAudioFormat(),
|
||||
mEngine.getChannelCount());
|
||||
|
||||
mEngine.setVoice(mMatchingVoice.name, gender);
|
||||
mEngine.setVoice(mMatchingVoice, null, gender, SpeechSynthesis.AGE_ANY);
|
||||
mEngine.setRate(rate);
|
||||
mEngine.setPitch(pitch);
|
||||
mEngine.synthesize(text, text.startsWith("<speak"));
|
||||
|
||||
Reference in New Issue
Block a user