Golang API for streaming zipformer2 CTC

.github/workflows/test-go.yaml

@@ -53,7 +53,7 @@ jobs:
           mkdir build
           cd build
           cmake -DCMAKE_C_COMPILER_LAUNCHER=ccache -DCMAKE_CXX_COMPILER_LAUNCHER=ccache -DBUILD_SHARED_LIBS=ON -DSHERPA_ONNX_ENABLE_PORTAUDIO=OFF -DSHERPA_ONNX_ENABLE_WEBSOCKET=OFF ..
-          make -j
+          make -j1
           cp -v _deps/onnxruntime-src/lib/libonnxruntime*dylib ./lib/
 
           cd ../scripts/go/_internal/
@@ -153,6 +153,14 @@ jobs:
 
           git lfs install
 
+          echo "Test zipformer2 CTC"
+          wget -qq https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-streaming-zipformer-ctc-multi-zh-hans-2023-12-13.tar.bz2
+          tar xvf sherpa-onnx-streaming-zipformer-ctc-multi-zh-hans-2023-12-13.tar.bz2
+          rm sherpa-onnx-streaming-zipformer-ctc-multi-zh-hans-2023-12-13.tar.bz2
+
+          ./run-zipformer2-ctc.sh
+          rm -rf sherpa-onnx-streaming-zipformer-ctc-multi-zh-hans-2023-12-13
+
           echo "Test transducer"
           git clone https://huggingface.co/csukuangfj/sherpa-onnx-streaming-zipformer-en-2023-06-26
           ./run-transducer.sh

go-api-examples/.gitignore

@@ -0,0 +1 @@
+!*.sh

go-api-examples/streaming-decode-files/main.go

@@ -22,6 +22,7 @@ func main() {
 	flag.StringVar(&config.ModelConfig.Transducer.Joiner, "joiner", "", "Path to the transducer joiner model")
 	flag.StringVar(&config.ModelConfig.Paraformer.Encoder, "paraformer-encoder", "", "Path to the paraformer encoder model")
 	flag.StringVar(&config.ModelConfig.Paraformer.Decoder, "paraformer-decoder", "", "Path to the paraformer decoder model")
+	flag.StringVar(&config.ModelConfig.Zipformer2Ctc.Model, "zipformer2-ctc", "", "Path to the zipformer2 CTC model")
 	flag.StringVar(&config.ModelConfig.Tokens, "tokens", "", "Path to the tokens file")
 	flag.IntVar(&config.ModelConfig.NumThreads, "num-threads", 1, "Number of threads for computing")
 	flag.IntVar(&config.ModelConfig.Debug, "debug", 0, "Whether to show debug message")

go-api-examples/streaming-decode-files/run-zipformer2-ctc.sh

@@ -0,0 +1,13 @@
+#!/usr/bin/env bash
+
+# Please refer to
+# https://k2-fsa.github.io/sherpa/onnx/pretrained_models/online-ctc/zipformer-ctc-models.html#sherpa-onnx-streaming-zipformer-ctc-multi-zh-hans-2023-12-13-chinese
+# to download the model
+# before you run this script.
+#
+# You can switch to a different online model if you need
+
+./streaming-decode-files \
+  --zipformer2-ctc ./sherpa-onnx-streaming-zipformer-ctc-multi-zh-hans-2023-12-13/ctc-epoch-20-avg-1-chunk-16-left-128.onnx \
+  --tokens ./sherpa-onnx-streaming-zipformer-ctc-multi-zh-hans-2023-12-13/tokens.txt \
+  ./sherpa-onnx-streaming-zipformer-ctc-multi-zh-hans-2023-12-13/test_wavs/DEV_T0000000000.wav

python-api-examples/offline-websocket-client-decode-files-paralell.py

@@ -129,7 +129,7 @@ async def run(
             await websocket.send(buf)
 
         decoding_results = await websocket.recv()
-        logging.info(f"{wave_filename}\n{json.loads(decoding_results)}")
+        logging.info(f"{wave_filename}\n{decoding_results}")
 
         # to signal that the client has sent all the data
         await websocket.send("Done")

scripts/go/_internal/streaming-decode-files/run-zipformer2-ctc.sh

@@ -0,0 +1 @@
+../../../../go-api-examples/streaming-decode-files/run-zipformer2-ctc.sh

scripts/go/sherpa_onnx.go

@@ -65,20 +65,28 @@ type OnlineParaformerModelConfig struct {
 	Decoder string // Path to the decoder model.
 }
 
+// Please refer to
+// https://k2-fsa.github.io/sherpa/onnx/pretrained_models/online-ctc/index.html
+// to download pre-trained models
+type OnlineZipformer2CtcModelConfig struct {
+	Model string // Path to the onnx model
+}
+
 // Configuration for online/streaming models
 //
 // Please refer to
 // https://k2-fsa.github.io/sherpa/onnx/pretrained_models/online-transducer/index.html
 // https://k2-fsa.github.io/sherpa/onnx/pretrained_models/online-paraformer/index.html
 // to download pre-trained models
 type OnlineModelConfig struct {
-	Transducer OnlineTransducerModelConfig
-	Paraformer OnlineParaformerModelConfig
-	Tokens     string // Path to tokens.txt
-	NumThreads int    // Number of threads to use for neural network computation
-	Provider   string // Optional. Valid values are: cpu, cuda, coreml
-	Debug      int    // 1 to show model meta information while loading it.
-	ModelType  string // Optional. You can specify it for faster model initialization
+	Transducer    OnlineTransducerModelConfig
+	Paraformer    OnlineParaformerModelConfig
+	Zipformer2Ctc OnlineZipformer2CtcModelConfig
+	Tokens        string // Path to tokens.txt
+	NumThreads    int    // Number of threads to use for neural network computation
+	Provider      string // Optional. Valid values are: cpu, cuda, coreml
+	Debug         int    // 1 to show model meta information while loading it.
+	ModelType     string // Optional. You can specify it for faster model initialization
 }
 
 // Configuration for the feature extractor
@@ -157,6 +165,9 @@ func NewOnlineRecognizer(config *OnlineRecognizerConfig) *OnlineRecognizer {
 	c.model_config.paraformer.decoder = C.CString(config.ModelConfig.Paraformer.Decoder)
 	defer C.free(unsafe.Pointer(c.model_config.paraformer.decoder))
 
+	c.model_config.zipformer2_ctc.model = C.CString(config.ModelConfig.Zipformer2Ctc.Model)
+	defer C.free(unsafe.Pointer(c.model_config.zipformer2_ctc.model))
+
 	c.model_config.tokens = C.CString(config.ModelConfig.Tokens)
 	defer C.free(unsafe.Pointer(c.model_config.tokens))