Use piper-phonemize to convert text to token IDs (#453)

swift-api-examples/SherpaOnnx.swift

@@ -578,6 +578,7 @@ func sherpaOnnxOfflineTtsVitsModelConfig(
   model: String,
   lexicon: String,
   tokens: String,
+  dataDir: String = "",
   noiseScale: Float = 0.667,
   noiseScaleW: Float = 0.8,
   lengthScale: Float = 1.0
@@ -586,6 +587,7 @@ func sherpaOnnxOfflineTtsVitsModelConfig(
     model: toCPointer(model),
     lexicon: toCPointer(lexicon),
     tokens: toCPointer(tokens),
+    data_dir: toCPointer(dataDir),
     noise_scale: noiseScale,
     noise_scale_w: noiseScaleW,
     length_scale: lengthScale)
@@ -607,11 +609,13 @@ func sherpaOnnxOfflineTtsModelConfig(
 
 func sherpaOnnxOfflineTtsConfig(
   model: SherpaOnnxOfflineTtsModelConfig,
-  ruleFsts: String = ""
+  ruleFsts: String = "",
+  maxNumSenetences: Int = 2
 ) -> SherpaOnnxOfflineTtsConfig {
   return SherpaOnnxOfflineTtsConfig(
     model: model,
-    rule_fsts: toCPointer(ruleFsts)
+    rule_fsts: toCPointer(ruleFsts),
+    max_num_sentences: Int32(maxNumSenetences)
   )
 }
 

swift-api-examples/run-tts.sh

@@ -7,17 +7,12 @@ if [ ! -d ../build-swift-macos ]; then
   exit 1
 fi
 
-if [ ! -d ./vits-vctk ]; then
-  echo "Please download the pre-trained model for testing."
-  echo "You can refer to"
-  echo ""
-  echo "https://k2-fsa.github.io/sherpa/onnx/tts/pretrained_models/vits.html#vctk-english-multi-speaker-109-speakers"
-  echo ""
-  echo "for help"
+if [ ! -d ./vits-piper-en_US-amy-low ]; then
+  echo "Download a pre-trained model for testing."
 
-  wget -q https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/vits-vctk.tar.bz2
-  tar xvf vits-vctk.tar.bz2
-  rm vits-vctk.tar.bz2
+  wget -q https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/vits-piper-en_US-amy-low.tar.bz2
+  tar xf vits-piper-en_US-amy-low.tar.bz2
+  rm vits-piper-en_US-amy-low.tar.bz2
 fi
 
 if [ ! -e ./tts ]; then

swift-api-examples/tts.swift

@@ -1,18 +1,20 @@
 func run() {
-  let model = "./vits-vctk/vits-vctk.onnx"
-  let lexicon = "./vits-vctk/lexicon.txt"
-  let tokens = "./vits-vctk/tokens.txt"
+  let model = "./vits-piper-en_US-amy-low/en_US-amy-low.onnx"
+  let tokens = "./vits-piper-en_US-amy-low/tokens.txt"
+  let dataDir = "./vits-piper-en_US-amy-low/espeak-ng-data"
   let vits = sherpaOnnxOfflineTtsVitsModelConfig(
     model: model,
-    lexicon: lexicon,
-    tokens: tokens
+    lexicon: "",
+    tokens: tokens,
+    dataDir: dataDir
   )
   let modelConfig = sherpaOnnxOfflineTtsModelConfig(vits: vits)
   var ttsConfig = sherpaOnnxOfflineTtsConfig(model: modelConfig)
 
   let tts = SherpaOnnxOfflineTtsWrapper(config: &ttsConfig)
 
-  let text = "How are you doing? Fantastic!"
+  let text =
+    "“Today as always, men fall into two groups: slaves and free men. Whoever does not have two-thirds of his day for himself, is a slave, whatever he may be: a statesman, a businessman, an official, or a scholar.”"
   let sid = 99
   let speed: Float = 1.0