Add Go API for KittenTTS (#2478)

@@ -311,6 +311,11 @@ jobs:
           go build
           go build
           ls -lh
           ls -lh
 
 
+          echo "Test kitten en"
+          ./run-kitten-en.sh
+          rm -rf kitten-*
+          ls -lh
+
           echo "Test kokoro zh+en"
           echo "Test kokoro zh+en"
           ./run-kokoro-zh-en.sh
           ./run-kokoro-zh-en.sh
           rm -rf kokoro-multi-*
           rm -rf kokoro-multi-*

@@ -41,6 +41,12 @@ func main() {
 	flag.StringVar(&config.Model.Kokoro.Lexicon, "kokoro-lexicon", "", "Path to lexicon files for Kokoro")
 	flag.StringVar(&config.Model.Kokoro.Lexicon, "kokoro-lexicon", "", "Path to lexicon files for Kokoro")
 	flag.Float32Var(&config.Model.Kokoro.LengthScale, "kokoro-length-scale", 1.0, "length_scale for Kokoro. small -> faster in speech speed; large -> slower")
 	flag.Float32Var(&config.Model.Kokoro.LengthScale, "kokoro-length-scale", 1.0, "length_scale for Kokoro. small -> faster in speech speed; large -> slower")
 
 
+	flag.StringVar(&config.Model.Kitten.Model, "kitten-model", "", "Path to the kitten ONNX model")
+	flag.StringVar(&config.Model.Kitten.Voices, "kitten-voices", "", "Path to voices.bin for kitten")
+	flag.StringVar(&config.Model.Kitten.Tokens, "kitten-tokens", "", "Path to tokens.txt for kitten")
+	flag.StringVar(&config.Model.Kitten.DataDir, "kitten-data-dir", "", "Path to espeak-ng-data for kitten")
+	flag.Float32Var(&config.Model.Kitten.LengthScale, "kitten-length-scale", 1.0, "length_scale for kitten. small -> faster in speech speed; large -> slower")
+
 	flag.IntVar(&config.Model.NumThreads, "num-threads", 1, "Number of threads for computing")
 	flag.IntVar(&config.Model.NumThreads, "num-threads", 1, "Number of threads for computing")
 	flag.IntVar(&config.Model.Debug, "debug", 0, "Whether to show debug message")
 	flag.IntVar(&config.Model.Debug, "debug", 0, "Whether to show debug message")
 	flag.StringVar(&config.Model.Provider, "provider", "cpu", "Provider to use")
 	flag.StringVar(&config.Model.Provider, "provider", "cpu", "Provider to use")

+#!/usr/bin/env bash
+
+set -ex
+
+if [ ! -f ./kitten-nano-en-v0_1-fp16/model.fp16.onnx ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/kitten-nano-en-v0_1-fp16.tar.bz2
+  tar xf kitten-nano-en-v0_1-fp16.tar.bz2
+  rm kitten-nano-en-v0_1-fp16.tar.bz2
+fi
+
+go mod tidy
+go build
+
+./non-streaming-tts \
+  --kitten-model=./kitten-nano-en-v0_1-fp16/model.fp16.onnx \
+  --kitten-voices=./kitten-nano-en-v0_1-fp16/voices.bin \
+  --kitten-tokens=./kitten-nano-en-v0_1-fp16/tokens.txt \
+  --kitten-data-dir=./kitten-nano-en-v0_1-fp16/espeak-ng-data \
+  --debug=1 \
+  --output-filename=./test-kitten-en.wav \
+  "Friends fell out often because life was changing so fast. The easiest thing in the world was to lose touch with someone."

@@ -41,6 +41,12 @@ func main() {
 	flag.StringVar(&config.Model.Kokoro.Lexicon, "kokoro-lexicon", "", "Path to lexicon files for Kokoro")
 	flag.StringVar(&config.Model.Kokoro.Lexicon, "kokoro-lexicon", "", "Path to lexicon files for Kokoro")
 	flag.Float32Var(&config.Model.Kokoro.LengthScale, "kokoro-length-scale", 1.0, "length_scale for Kokoro. small -> faster in speech speed; large -> slower")
 	flag.Float32Var(&config.Model.Kokoro.LengthScale, "kokoro-length-scale", 1.0, "length_scale for Kokoro. small -> faster in speech speed; large -> slower")
 
 
+	flag.StringVar(&config.Model.Kitten.Model, "kitten-model", "", "Path to the kitten ONNX model")
+	flag.StringVar(&config.Model.Kitten.Voices, "kitten-voices", "", "Path to voices.bin for kitten")
+	flag.StringVar(&config.Model.Kitten.Tokens, "kitten-tokens", "", "Path to tokens.txt for kitten")
+	flag.StringVar(&config.Model.Kitten.DataDir, "kitten-data-dir", "", "Path to espeak-ng-data for kitten")
+	flag.Float32Var(&config.Model.Kitten.LengthScale, "kitten-length-scale", 1.0, "length_scale for kitten. small -> faster in speech speed; large -> slower")
+
 	flag.IntVar(&config.Model.NumThreads, "num-threads", 1, "Number of threads for computing")
 	flag.IntVar(&config.Model.NumThreads, "num-threads", 1, "Number of threads for computing")
 	flag.IntVar(&config.Model.Debug, "debug", 0, "Whether to show debug message")
 	flag.IntVar(&config.Model.Debug, "debug", 0, "Whether to show debug message")
 	flag.StringVar(&config.Model.Provider, "provider", "cpu", "Provider to use")
 	flag.StringVar(&config.Model.Provider, "provider", "cpu", "Provider to use")

+#!/usr/bin/env bash
+
+set -ex
+
+if [ ! -f ./kitten-nano-en-v0_1-fp16/model.fp16.onnx ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/kitten-nano-en-v0_1-fp16.tar.bz2
+  tar xf kitten-nano-en-v0_1-fp16.tar.bz2
+  rm kitten-nano-en-v0_1-fp16.tar.bz2
+fi
+
+go mod tidy
+go build
+
+./offline-tts-play \
+  --kitten-model=./kitten-nano-en-v0_1-fp16/model.fp16.onnx \
+  --kitten-voices=./kitten-nano-en-v0_1-fp16/voices.bin \
+  --kitten-tokens=./kitten-nano-en-v0_1-fp16/tokens.txt \
+  --kitten-data-dir=./kitten-nano-en-v0_1-fp16/espeak-ng-data \
+  --debug=1 \
+  "Friends fell out often because life was changing so fast. The easiest thing in the world was to lose touch with someone."

 module non-streaming-tts
 module non-streaming-tts
 
 
 go 1.17
 go 1.17
-
-replace github.com/k2-fsa/sherpa-onnx-go/sherpa_onnx => ../

 module offline-tts-play
 module offline-tts-play
 
 
 go 1.17
 go 1.17
-
-replace github.com/k2-fsa/sherpa-onnx-go/sherpa_onnx => ../
-
-require (
-	github.com/k2-fsa/sherpa-onnx-go/sherpa_onnx v0.0.0-00010101000000-000000000000
-	github.com/spf13/pflag v1.0.6
-)

-../../../../go-api-examples/non-streaming-tts/main.go
+../../../../go-api-examples/offline-tts-play/main.go

+../../../../go-api-examples/offline-tts-play/run-kitten-en.sh

@@ -921,10 +921,19 @@ type OfflineTtsKokoroModelConfig struct {
 	LengthScale float32 // Please use 1.0 in general. Smaller -> Faster speech speed. Larger -> Slower speech speed
 	LengthScale float32 // Please use 1.0 in general. Smaller -> Faster speech speed. Larger -> Slower speech speed
 }
 }
 
 
+type OfflineTtsKittenModelConfig struct {
+	Model       string  // Path to the model for kitten
+	Voices      string  // Path to the voices.bin for kitten
+	Tokens      string  // Path to tokens.txt
+	DataDir     string  // Path to espeak-ng-data directory
+	LengthScale float32 // Please use 1.0 in general. Smaller -> Faster speech speed. Larger -> Slower speech speed
+}
+
 type OfflineTtsModelConfig struct {
 type OfflineTtsModelConfig struct {
 	Vits   OfflineTtsVitsModelConfig
 	Vits   OfflineTtsVitsModelConfig
 	Matcha OfflineTtsMatchaModelConfig
 	Matcha OfflineTtsMatchaModelConfig
 	Kokoro OfflineTtsKokoroModelConfig
 	Kokoro OfflineTtsKokoroModelConfig
+	Kitten OfflineTtsKittenModelConfig
 
 
 	// Number of threads to use for neural network computation
 	// Number of threads to use for neural network computation
 	NumThreads int
 	NumThreads int
@@ -1072,6 +1081,21 @@ func NewOfflineTts(config *OfflineTtsConfig) *OfflineTts {
 
 
 	c.model.kokoro.length_scale = C.float(config.Model.Kokoro.LengthScale)
 	c.model.kokoro.length_scale = C.float(config.Model.Kokoro.LengthScale)
 
 
+	// kitten
+	c.model.kitten.model = C.CString(config.Model.Kitten.Model)
+	defer C.free(unsafe.Pointer(c.model.kitten.model))
+
+	c.model.kitten.voices = C.CString(config.Model.Kitten.Voices)
+	defer C.free(unsafe.Pointer(c.model.kitten.voices))
+
+	c.model.kitten.tokens = C.CString(config.Model.Kitten.Tokens)
+	defer C.free(unsafe.Pointer(c.model.kitten.tokens))
+
+	c.model.kitten.data_dir = C.CString(config.Model.Kitten.DataDir)
+	defer C.free(unsafe.Pointer(c.model.kitten.data_dir))
+
+	c.model.kitten.length_scale = C.float(config.Model.Kitten.LengthScale)
+
 	c.model.num_threads = C.int(config.Model.NumThreads)
 	c.model.num_threads = C.int(config.Model.NumThreads)
 	c.model.debug = C.int(config.Model.Debug)
 	c.model.debug = C.int(config.Model.Debug)