Add go-api-examples (#219)
This commit is contained in:
Fangjun Kuang
2
go-api-examples/non-streaming-decode-files/.gitignore
vendored
Normal file
2
go-api-examples/non-streaming-decode-files/.gitignore
vendored
Normal file
@@ -0,0 +1,2 @@
|
||||
non-streaming-decode-files
|
||||
sherpa-onnx-zipformer-en-2023-06-26
|
||||
17
go-api-examples/non-streaming-decode-files/go.mod
Normal file
17
go-api-examples/non-streaming-decode-files/go.mod
Normal file
@@ -0,0 +1,17 @@
|
||||
module non-streaming-decode-files
|
||||
|
||||
go 1.20
|
||||
|
||||
require (
|
||||
github.com/k2-fsa/sherpa-onnx-go v1.5.3-alpha.8
|
||||
github.com/spf13/pflag v1.0.5
|
||||
github.com/youpy/go-wav v0.3.2
|
||||
)
|
||||
|
||||
require (
|
||||
github.com/k2-fsa/sherpa-onnx-go-linux v1.5.3-alpha.6 // indirect
|
||||
github.com/k2-fsa/sherpa-onnx-go-macos v1.5.3-alpha.4 // indirect
|
||||
github.com/k2-fsa/sherpa-onnx-go-windows v1.5.3-alpha.5 // indirect
|
||||
github.com/youpy/go-riff v0.1.0 // indirect
|
||||
github.com/zaf/g711 v0.0.0-20190814101024-76a4a538f52b // indirect
|
||||
)
|
||||
33
go-api-examples/non-streaming-decode-files/go.sum
Normal file
33
go-api-examples/non-streaming-decode-files/go.sum
Normal file
@@ -0,0 +1,33 @@
|
||||
github.com/davecgh/go-spew v1.1.0 h1:ZDRjVQ15GmhC3fiQ8ni8+OwkZQO4DARzQgrnXU1Liz8=
|
||||
github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
|
||||
github.com/google/go-cmp v0.5.6 h1:BKbKCqvP6I+rmFHt06ZmyQtvB8xAkWdhFyr0ZUNZcxQ=
|
||||
github.com/google/go-cmp v0.5.6/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
|
||||
github.com/k2-fsa/sherpa-onnx-go v1.5.3-alpha.8 h1:BXc31pWwd7CJLM6m9HavxqiyYJdN3Jc9I26pd4x+JHE=
|
||||
github.com/k2-fsa/sherpa-onnx-go v1.5.3-alpha.8/go.mod h1:kszL/pwg9XTpRGi1AYW/aSwdhRBqb6LN0SjXR0jnsBo=
|
||||
github.com/k2-fsa/sherpa-onnx-go-linux v1.5.3-alpha.6 h1:o0+l4Wr3IWkWH+kdt8ZZP55L8mRSpW7h1KzvXcfx9FM=
|
||||
github.com/k2-fsa/sherpa-onnx-go-linux v1.5.3-alpha.6/go.mod h1:9kU02PSdDdzBApwIDBmE2jWS54WvZbRafOEvW/PVLxE=
|
||||
github.com/k2-fsa/sherpa-onnx-go-macos v1.5.3-alpha.4 h1:Xri23R3+tQFkNwO6zVxqZbjMRJP40z7JtoSqW6UP6sM=
|
||||
github.com/k2-fsa/sherpa-onnx-go-macos v1.5.3-alpha.4/go.mod h1:a+AJZKNQkFO+JyzGkHySysYfBzzdcoJI5ITFsnhVcmo=
|
||||
github.com/k2-fsa/sherpa-onnx-go-windows v1.5.3-alpha.5 h1:OR1LLoptR8W35j5u06KrYUblG35B83HHCPWrVK31uHY=
|
||||
github.com/k2-fsa/sherpa-onnx-go-windows v1.5.3-alpha.5/go.mod h1:ermMOETZUv0nM7MmnXWqeHREMR6zQXBhJpEP2fbHIZo=
|
||||
github.com/pkg/errors v0.9.1 h1:FEBLx1zS214owpjy7qsBeixbURkuhQAwrK5UwLGTwt4=
|
||||
github.com/pkg/errors v0.9.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
|
||||
github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
|
||||
github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
|
||||
github.com/spf13/pflag v1.0.5 h1:iy+VFUOCP1a+8yFto/drg2CJ5u0yRoB7fZw3DKv/JXA=
|
||||
github.com/spf13/pflag v1.0.5/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
|
||||
github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
|
||||
github.com/stretchr/testify v1.6.1 h1:hDPOHmpOpP40lSULcqw7IrRb/u7w6RpDC9399XyoNd0=
|
||||
github.com/stretchr/testify v1.6.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
|
||||
github.com/youpy/go-riff v0.1.0 h1:vZO/37nI4tIET8tQI0Qn0Y79qQh99aEpponTPiPut7k=
|
||||
github.com/youpy/go-riff v0.1.0/go.mod h1:83nxdDV4Z9RzrTut9losK7ve4hUnxUR8ASSz4BsKXwQ=
|
||||
github.com/youpy/go-wav v0.3.2 h1:NLM8L/7yZ0Bntadw/0h95OyUsen+DQIVf9gay+SUsMU=
|
||||
github.com/youpy/go-wav v0.3.2/go.mod h1:0FCieAXAeSdcxFfwLpRuEo0PFmAoc+8NU34h7TUvk50=
|
||||
github.com/zaf/g711 v0.0.0-20190814101024-76a4a538f52b h1:QqixIpc5WFIqTLxB3Hq8qs0qImAgBdq0p6rq2Qdl634=
|
||||
github.com/zaf/g711 v0.0.0-20190814101024-76a4a538f52b/go.mod h1:T2h1zV50R/q0CVYnsQOQ6L7P4a2ZxH47ixWcMXFGyx8=
|
||||
golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
|
||||
gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
|
||||
gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c h1:dUUwHk2QECo/6vqA44rthZ8ie2QXMNeKRTHCNY2nXvo=
|
||||
gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
|
||||
gotest.tools v2.2.0+incompatible h1:VsBPFP1AI068pPrMxtb/S8Zkgf9xEmTLJjfM+P5UIEo=
|
||||
gotest.tools v2.2.0+incompatible/go.mod h1:DsYFclhRJ6vuDpmuTbkuFWG+y2sxOXAzmJt81HFBacw=
|
||||
120
go-api-examples/non-streaming-decode-files/main.go
Normal file
120
go-api-examples/non-streaming-decode-files/main.go
Normal file
@@ -0,0 +1,120 @@
|
||||
package main
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"encoding/binary"
|
||||
sherpa "github.com/k2-fsa/sherpa-onnx-go/sherpa_onnx"
|
||||
flag "github.com/spf13/pflag"
|
||||
"github.com/youpy/go-wav"
|
||||
"os"
|
||||
"strings"
|
||||
|
||||
"log"
|
||||
)
|
||||
|
||||
func main() {
|
||||
|
||||
log.SetFlags(log.LstdFlags | log.Lmicroseconds)
|
||||
|
||||
config := sherpa.OfflineRecognizerConfig{}
|
||||
config.FeatConfig = sherpa.FeatureConfig{SampleRate: 16000, FeatureDim: 80}
|
||||
|
||||
flag.StringVar(&config.ModelConfig.Transducer.Encoder, "encoder", "", "Path to the encoder model")
|
||||
flag.StringVar(&config.ModelConfig.Transducer.Decoder, "decoder", "", "Path to the decoder model")
|
||||
flag.StringVar(&config.ModelConfig.Transducer.Joiner, "joiner", "", "Path to the joiner model")
|
||||
flag.StringVar(&config.ModelConfig.Paraformer.Model, "paraformer", "", "Path to the paraformer model")
|
||||
flag.StringVar(&config.ModelConfig.NemoCTC.Model, "nemo-ctc", "", "Path to the NeMo 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")
|
||||
flag.StringVar(&config.ModelConfig.ModelType, "model-type", "", "Optional. Used for loading the model in a faster way")
|
||||
flag.StringVar(&config.ModelConfig.Provider, "provider", "cpu", "Provider to use")
|
||||
flag.StringVar(&config.LmConfig.Model, "lm-model", "", "Optional. Path to the LM model")
|
||||
flag.Float32Var(&config.LmConfig.Scale, "lm-scale", 1.0, "Optional. Scale for the LM model")
|
||||
|
||||
flag.StringVar(&config.DecodingMethod, "decoding-method", "greedy_search", "Decoding method. Possible values: greedy_search, modified_beam_search")
|
||||
flag.IntVar(&config.MaxActivePaths, "max-active-paths", 4, "Used only when --decoding-method is modified_beam_search")
|
||||
|
||||
flag.Parse()
|
||||
|
||||
if len(flag.Args()) != 1 {
|
||||
log.Fatalf("Please provide one wave file")
|
||||
}
|
||||
|
||||
log.Println("Reading", flag.Arg(0))
|
||||
|
||||
samples, sampleRate := readWave(flag.Arg(0))
|
||||
|
||||
log.Println("Initializing recognizer (may take several seconds)")
|
||||
recognizer := sherpa.NewOfflineRecognizer(&config)
|
||||
log.Println("Recognizer created!")
|
||||
defer sherpa.DeleteOfflineRecognizer(recognizer)
|
||||
|
||||
log.Println("Start decoding!")
|
||||
stream := sherpa.NewOfflineStream(recognizer)
|
||||
defer sherpa.DeleteOfflineStream(stream)
|
||||
|
||||
stream.AcceptWaveform(sampleRate, samples)
|
||||
|
||||
recognizer.Decode(stream)
|
||||
log.Println("Decoding done!")
|
||||
result := stream.GetResult()
|
||||
|
||||
log.Println(strings.ToLower(result.Text))
|
||||
log.Printf("Wave duration: %v seconds", float32(len(samples))/float32(sampleRate))
|
||||
}
|
||||
|
||||
func readWave(filename string) (samples []float32, sampleRate int) {
|
||||
file, _ := os.Open(filename)
|
||||
defer file.Close()
|
||||
|
||||
reader := wav.NewReader(file)
|
||||
format, err := reader.Format()
|
||||
if err != nil {
|
||||
log.Fatalf("Failed to read wave format")
|
||||
}
|
||||
|
||||
if format.AudioFormat != 1 {
|
||||
log.Fatalf("Support only PCM format. Given: %v\n", format.AudioFormat)
|
||||
}
|
||||
|
||||
if format.NumChannels != 1 {
|
||||
log.Fatalf("Support only 1 channel wave file. Given: %v\n", format.NumChannels)
|
||||
}
|
||||
|
||||
if format.BitsPerSample != 16 {
|
||||
log.Fatalf("Support only 16-bit per sample. Given: %v\n", format.BitsPerSample)
|
||||
}
|
||||
|
||||
reader.Duration() // so that it initializes reader.Size
|
||||
|
||||
buf := make([]byte, reader.Size)
|
||||
n, err := reader.Read(buf)
|
||||
if n != int(reader.Size) {
|
||||
log.Fatalf("Failed to read %v bytes. Returned %v bytes\n", reader.Size, n)
|
||||
}
|
||||
|
||||
samples = samplesInt16ToFloat(buf)
|
||||
sampleRate = int(format.SampleRate)
|
||||
|
||||
return
|
||||
}
|
||||
|
||||
func samplesInt16ToFloat(inSamples []byte) []float32 {
|
||||
numSamples := len(inSamples) / 2
|
||||
outSamples := make([]float32, numSamples)
|
||||
|
||||
for i := 0; i != numSamples; i++ {
|
||||
s := inSamples[i*2 : (i+1)*2]
|
||||
|
||||
var s16 int16
|
||||
buf := bytes.NewReader(s)
|
||||
err := binary.Read(buf, binary.LittleEndian, &s16)
|
||||
if err != nil {
|
||||
log.Fatal("Failed to parse 16-bit sample")
|
||||
}
|
||||
outSamples[i] = float32(s16) / 32768
|
||||
}
|
||||
|
||||
return outSamples
|
||||
}
|
||||
15
go-api-examples/non-streaming-decode-files/run-nemo-ctc.sh
Executable file
15
go-api-examples/non-streaming-decode-files/run-nemo-ctc.sh
Executable file
@@ -0,0 +1,15 @@
|
||||
#!/usr/bin/env bash
|
||||
|
||||
# Please refer to
|
||||
# https://k2-fsa.github.io/sherpa/onnx/pretrained_models/offline-ctc/nemo/english.html#stt-en-conformer-ctc-medium
|
||||
# to download the model
|
||||
# before you run this script.
|
||||
#
|
||||
# You can switch to a different online model if you need
|
||||
|
||||
./non-streaming-decode-files \
|
||||
--nemo-ctc ./sherpa-onnx-nemo-ctc-en-conformer-medium/model.onnx \
|
||||
--tokens ./sherpa-onnx-nemo-ctc-en-conformer-medium/tokens.txt \
|
||||
--model-type nemo_ctc \
|
||||
--debug 0 \
|
||||
./sherpa-onnx-nemo-ctc-en-conformer-medium/test_wavs/0.wav
|
||||
15
go-api-examples/non-streaming-decode-files/run-paraformer.sh
Executable file
15
go-api-examples/non-streaming-decode-files/run-paraformer.sh
Executable file
@@ -0,0 +1,15 @@
|
||||
#!/usr/bin/env bash
|
||||
|
||||
# Please refer to
|
||||
# https://k2-fsa.github.io/sherpa/onnx/pretrained_models/offline-paraformer/paraformer-models.html#csukuangfj-sherpa-onnx-paraformer-zh-2023-03-28-chinese
|
||||
# to download the model
|
||||
# before you run this script.
|
||||
#
|
||||
# You can switch to a different online model if you need
|
||||
|
||||
./non-streaming-decode-files \
|
||||
--paraformer ./sherpa-onnx-paraformer-zh-2023-03-28/model.int8.onnx \
|
||||
--tokens ./sherpa-onnx-paraformer-zh-2023-03-28/tokens.txt \
|
||||
--model-type paraformer \
|
||||
--debug 0 \
|
||||
./sherpa-onnx-paraformer-zh-2023-03-28/test_wavs/0.wav
|
||||
17
go-api-examples/non-streaming-decode-files/run-transducer.sh
Executable file
17
go-api-examples/non-streaming-decode-files/run-transducer.sh
Executable file
@@ -0,0 +1,17 @@
|
||||
#!/usr/bin/env bash
|
||||
|
||||
# Please refer to
|
||||
# https://k2-fsa.github.io/sherpa/onnx/pretrained_models/offline-transducer/zipformer-transducer-models.html#csukuangfj-sherpa-onnx-zipformer-en-2023-06-26-english
|
||||
# to download the model
|
||||
# before you run this script.
|
||||
#
|
||||
# You can switch to a different online model if you need
|
||||
|
||||
./non-streaming-decode-files \
|
||||
--encoder ./sherpa-onnx-zipformer-en-2023-06-26/encoder-epoch-99-avg-1.onnx \
|
||||
--decoder ./sherpa-onnx-zipformer-en-2023-06-26/decoder-epoch-99-avg-1.onnx \
|
||||
--joiner ./sherpa-onnx-zipformer-en-2023-06-26/joiner-epoch-99-avg-1.onnx \
|
||||
--tokens ./sherpa-onnx-zipformer-en-2023-06-26/tokens.txt \
|
||||
--model-type transducer \
|
||||
--debug 0 \
|
||||
./sherpa-onnx-zipformer-en-2023-06-26/test_wavs/0.wav
|
||||
1
go-api-examples/real-time-speech-recognition-from-microphone/.gitignore
vendored
Normal file
1
go-api-examples/real-time-speech-recognition-from-microphone/.gitignore
vendored
Normal file
@@ -0,0 +1 @@
|
||||
real-time-speech-recognition-from-microphone
|
||||
@@ -0,0 +1,23 @@
|
||||
# Introduction
|
||||
|
||||
This examples shows how to use the golang package of [sherpa-onnx][sherpa-onnx]
|
||||
for real-time speech recognition from microphone.
|
||||
|
||||
It uses <https://github.com/gordonklaus/portaudio>
|
||||
to read the microphone and you have to install `portaudio` first.
|
||||
|
||||
On macOS, you can use
|
||||
|
||||
```
|
||||
brew install portaudio
|
||||
```
|
||||
|
||||
and it will install `portaudio` into `/usr/local/Cellar/portaudio/19.7.0`.
|
||||
You need to set the following environment variable
|
||||
```
|
||||
export PKG_CONFIG_PATH=/usr/local/Cellar/portaudio/19.7.0
|
||||
```
|
||||
|
||||
so that `pkg-config --cflags --libs portaudio-2.0` can run successfully.
|
||||
|
||||
[sherpa-onnx]: https://github.com/k2-fsa/sherpa-onnx
|
||||
@@ -0,0 +1,15 @@
|
||||
module real-time-speech-recognition-from-microphone
|
||||
|
||||
go 1.20
|
||||
|
||||
require (
|
||||
github.com/gordonklaus/portaudio v0.0.0-20230709114228-aafa478834f5
|
||||
github.com/k2-fsa/sherpa-onnx-go v1.5.3-alpha.8
|
||||
github.com/spf13/pflag v1.0.5
|
||||
)
|
||||
|
||||
require (
|
||||
github.com/k2-fsa/sherpa-onnx-go-linux v1.5.3-alpha.6 // indirect
|
||||
github.com/k2-fsa/sherpa-onnx-go-macos v1.5.3-alpha.4 // indirect
|
||||
github.com/k2-fsa/sherpa-onnx-go-windows v1.5.3-alpha.5 // indirect
|
||||
)
|
||||
@@ -0,0 +1,12 @@
|
||||
github.com/gordonklaus/portaudio v0.0.0-20230709114228-aafa478834f5 h1:5AlozfqaVjGYGhms2OsdUyfdJME76E6rx5MdGpjzZpc=
|
||||
github.com/gordonklaus/portaudio v0.0.0-20230709114228-aafa478834f5/go.mod h1:WY8R6YKlI2ZI3UyzFk7P6yGSuS+hFwNtEzrexRyD7Es=
|
||||
github.com/k2-fsa/sherpa-onnx-go v1.5.3-alpha.8 h1:BXc31pWwd7CJLM6m9HavxqiyYJdN3Jc9I26pd4x+JHE=
|
||||
github.com/k2-fsa/sherpa-onnx-go v1.5.3-alpha.8/go.mod h1:kszL/pwg9XTpRGi1AYW/aSwdhRBqb6LN0SjXR0jnsBo=
|
||||
github.com/k2-fsa/sherpa-onnx-go-linux v1.5.3-alpha.6 h1:o0+l4Wr3IWkWH+kdt8ZZP55L8mRSpW7h1KzvXcfx9FM=
|
||||
github.com/k2-fsa/sherpa-onnx-go-linux v1.5.3-alpha.6/go.mod h1:9kU02PSdDdzBApwIDBmE2jWS54WvZbRafOEvW/PVLxE=
|
||||
github.com/k2-fsa/sherpa-onnx-go-macos v1.5.3-alpha.4 h1:Xri23R3+tQFkNwO6zVxqZbjMRJP40z7JtoSqW6UP6sM=
|
||||
github.com/k2-fsa/sherpa-onnx-go-macos v1.5.3-alpha.4/go.mod h1:a+AJZKNQkFO+JyzGkHySysYfBzzdcoJI5ITFsnhVcmo=
|
||||
github.com/k2-fsa/sherpa-onnx-go-windows v1.5.3-alpha.5 h1:OR1LLoptR8W35j5u06KrYUblG35B83HHCPWrVK31uHY=
|
||||
github.com/k2-fsa/sherpa-onnx-go-windows v1.5.3-alpha.5/go.mod h1:ermMOETZUv0nM7MmnXWqeHREMR6zQXBhJpEP2fbHIZo=
|
||||
github.com/spf13/pflag v1.0.5 h1:iy+VFUOCP1a+8yFto/drg2CJ5u0yRoB7fZw3DKv/JXA=
|
||||
github.com/spf13/pflag v1.0.5/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
|
||||
@@ -0,0 +1,110 @@
|
||||
package main
|
||||
|
||||
import (
|
||||
"github.com/gordonklaus/portaudio"
|
||||
sherpa "github.com/k2-fsa/sherpa-onnx-go/sherpa_onnx"
|
||||
flag "github.com/spf13/pflag"
|
||||
"strings"
|
||||
|
||||
"fmt"
|
||||
"log"
|
||||
)
|
||||
|
||||
func main() {
|
||||
err := portaudio.Initialize()
|
||||
if err != nil {
|
||||
log.Fatalf("Unable to initialize portaudio: %v\n", err)
|
||||
}
|
||||
defer portaudio.Terminate()
|
||||
|
||||
default_device, err := portaudio.DefaultInputDevice()
|
||||
if err != nil {
|
||||
log.Fatal("Failed to get default input device: %v\n", err)
|
||||
}
|
||||
fmt.Printf("Select default input device: %s\n", default_device.Name)
|
||||
param := portaudio.StreamParameters{}
|
||||
param.Input.Device = default_device
|
||||
param.Input.Channels = 1
|
||||
param.Input.Latency = default_device.DefaultLowInputLatency
|
||||
|
||||
param.SampleRate = 16000
|
||||
param.FramesPerBuffer = 0
|
||||
param.Flags = portaudio.ClipOff
|
||||
|
||||
config := sherpa.OnlineRecognizerConfig{}
|
||||
config.FeatConfig = sherpa.FeatureConfig{SampleRate: 16000, FeatureDim: 80}
|
||||
|
||||
flag.StringVar(&config.ModelConfig.Encoder, "encoder", "", "Path to the encoder model")
|
||||
flag.StringVar(&config.ModelConfig.Decoder, "decoder", "", "Path to the decoder model")
|
||||
flag.StringVar(&config.ModelConfig.Joiner, "joiner", "", "Path to the joiner 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")
|
||||
flag.StringVar(&config.ModelConfig.ModelType, "model-type", "", "Optional. Used for loading the model in a faster way")
|
||||
flag.StringVar(&config.ModelConfig.Provider, "provider", "cpu", "Provider to use")
|
||||
flag.StringVar(&config.DecodingMethod, "decoding-method", "greedy_search", "Decoding method. Possible values: greedy_search, modified_beam_search")
|
||||
flag.IntVar(&config.MaxActivePaths, "max-active-paths", 4, "Used only when --decoding-method is modified_beam_search")
|
||||
flag.IntVar(&config.EnableEndpoint, "enable-endpoint", 1, "Whether to enable endpoint")
|
||||
flag.Float32Var(&config.Rule1MinTrailingSilence, "rule1-min-trailing-silence", 2.4, "Threshold for rule1")
|
||||
flag.Float32Var(&config.Rule2MinTrailingSilence, "rule2-min-trailing-silence", 1.2, "Threshold for rule2")
|
||||
flag.Float32Var(&config.Rule3MinUtteranceLength, "rule3-min-utterance-length", 20, "Threshold for rule3")
|
||||
|
||||
flag.Parse()
|
||||
|
||||
log.Println("Initializing recognizer (may take several seconds)")
|
||||
recognizer := sherpa.NewOnlineRecognizer(&config)
|
||||
log.Println("Recognizer created!")
|
||||
defer sherpa.DeleteOnlineRecognizer(recognizer)
|
||||
|
||||
stream := sherpa.NewOnlineStream(recognizer)
|
||||
|
||||
// you can choose another value for 0.1 if you want
|
||||
samplesPerCall := int32(param.SampleRate * 0.1) // 0.1 second
|
||||
|
||||
samples := make([]float32, samplesPerCall)
|
||||
s, err := portaudio.OpenStream(param, samples)
|
||||
if err != nil {
|
||||
log.Fatalf("Failed to open the stream")
|
||||
}
|
||||
defer s.Close()
|
||||
chk(s.Start())
|
||||
|
||||
var last_text string
|
||||
|
||||
segment_idx := 0
|
||||
|
||||
fmt.Println("Started! Please speak")
|
||||
|
||||
for {
|
||||
chk(s.Read())
|
||||
stream.AcceptWaveform(int(param.SampleRate), samples)
|
||||
|
||||
for recognizer.IsReady(stream) {
|
||||
recognizer.Decode(stream)
|
||||
}
|
||||
|
||||
text := recognizer.GetResult(stream).Text
|
||||
if len(text) != 0 && last_text != text {
|
||||
last_text = strings.ToLower(text)
|
||||
fmt.Printf("\r%d: %s", segment_idx, last_text)
|
||||
}
|
||||
|
||||
if recognizer.IsEndpoint(stream) {
|
||||
if len(text) != 0 {
|
||||
segment_idx++
|
||||
fmt.Println()
|
||||
}
|
||||
recognizer.Reset(stream)
|
||||
}
|
||||
}
|
||||
|
||||
chk(s.Stop())
|
||||
return
|
||||
|
||||
}
|
||||
|
||||
func chk(err error) {
|
||||
if err != nil {
|
||||
panic(err)
|
||||
}
|
||||
}
|
||||
15
go-api-examples/real-time-speech-recognition-from-microphone/run.sh
Executable file
15
go-api-examples/real-time-speech-recognition-from-microphone/run.sh
Executable file
@@ -0,0 +1,15 @@
|
||||
#!/usr/bin/env bash
|
||||
|
||||
# Please refer to
|
||||
# https://k2-fsa.github.io/sherpa/onnx/pretrained_models/online-transducer/zipformer-transducer-models.html#pkufool-icefall-asr-zipformer-streaming-wenetspeech-20230615-chinese
|
||||
# to download the model
|
||||
# before you run this script.
|
||||
#
|
||||
# You can switch to different online models if you need
|
||||
|
||||
./real-time-speech-recognition-from-microphone \
|
||||
--encoder ./icefall-asr-zipformer-streaming-wenetspeech-20230615/exp/encoder-epoch-12-avg-4-chunk-16-left-128.onnx \
|
||||
--decoder ./icefall-asr-zipformer-streaming-wenetspeech-20230615/exp/decoder-epoch-12-avg-4-chunk-16-left-128.onnx \
|
||||
--joiner ./icefall-asr-zipformer-streaming-wenetspeech-20230615/exp/joiner-epoch-12-avg-4-chunk-16-left-128.onnx \
|
||||
--tokens ./icefall-asr-zipformer-streaming-wenetspeech-20230615/data/lang_char/tokens.txt \
|
||||
--model-type zipformer2
|
||||
1
go-api-examples/streaming-decode-files/.gitignore
vendored
Normal file
1
go-api-examples/streaming-decode-files/.gitignore
vendored
Normal file
@@ -0,0 +1 @@
|
||||
streaming-decode-files
|
||||
17
go-api-examples/streaming-decode-files/go.mod
Normal file
17
go-api-examples/streaming-decode-files/go.mod
Normal file
@@ -0,0 +1,17 @@
|
||||
module streaming-decode-files
|
||||
|
||||
go 1.20
|
||||
|
||||
require (
|
||||
github.com/k2-fsa/sherpa-onnx-go v1.5.3-alpha.8
|
||||
github.com/spf13/pflag v1.0.5
|
||||
github.com/youpy/go-wav v0.3.2
|
||||
)
|
||||
|
||||
require (
|
||||
github.com/k2-fsa/sherpa-onnx-go-linux v1.5.3-alpha.6 // indirect
|
||||
github.com/k2-fsa/sherpa-onnx-go-macos v1.5.3-alpha.4 // indirect
|
||||
github.com/k2-fsa/sherpa-onnx-go-windows v1.5.3-alpha.5 // indirect
|
||||
github.com/youpy/go-riff v0.1.0 // indirect
|
||||
github.com/zaf/g711 v0.0.0-20190814101024-76a4a538f52b // indirect
|
||||
)
|
||||
33
go-api-examples/streaming-decode-files/go.sum
Normal file
33
go-api-examples/streaming-decode-files/go.sum
Normal file
@@ -0,0 +1,33 @@
|
||||
github.com/davecgh/go-spew v1.1.0 h1:ZDRjVQ15GmhC3fiQ8ni8+OwkZQO4DARzQgrnXU1Liz8=
|
||||
github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
|
||||
github.com/google/go-cmp v0.5.6 h1:BKbKCqvP6I+rmFHt06ZmyQtvB8xAkWdhFyr0ZUNZcxQ=
|
||||
github.com/google/go-cmp v0.5.6/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
|
||||
github.com/k2-fsa/sherpa-onnx-go v1.5.3-alpha.8 h1:BXc31pWwd7CJLM6m9HavxqiyYJdN3Jc9I26pd4x+JHE=
|
||||
github.com/k2-fsa/sherpa-onnx-go v1.5.3-alpha.8/go.mod h1:kszL/pwg9XTpRGi1AYW/aSwdhRBqb6LN0SjXR0jnsBo=
|
||||
github.com/k2-fsa/sherpa-onnx-go-linux v1.5.3-alpha.6 h1:o0+l4Wr3IWkWH+kdt8ZZP55L8mRSpW7h1KzvXcfx9FM=
|
||||
github.com/k2-fsa/sherpa-onnx-go-linux v1.5.3-alpha.6/go.mod h1:9kU02PSdDdzBApwIDBmE2jWS54WvZbRafOEvW/PVLxE=
|
||||
github.com/k2-fsa/sherpa-onnx-go-macos v1.5.3-alpha.4 h1:Xri23R3+tQFkNwO6zVxqZbjMRJP40z7JtoSqW6UP6sM=
|
||||
github.com/k2-fsa/sherpa-onnx-go-macos v1.5.3-alpha.4/go.mod h1:a+AJZKNQkFO+JyzGkHySysYfBzzdcoJI5ITFsnhVcmo=
|
||||
github.com/k2-fsa/sherpa-onnx-go-windows v1.5.3-alpha.5 h1:OR1LLoptR8W35j5u06KrYUblG35B83HHCPWrVK31uHY=
|
||||
github.com/k2-fsa/sherpa-onnx-go-windows v1.5.3-alpha.5/go.mod h1:ermMOETZUv0nM7MmnXWqeHREMR6zQXBhJpEP2fbHIZo=
|
||||
github.com/pkg/errors v0.9.1 h1:FEBLx1zS214owpjy7qsBeixbURkuhQAwrK5UwLGTwt4=
|
||||
github.com/pkg/errors v0.9.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
|
||||
github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
|
||||
github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
|
||||
github.com/spf13/pflag v1.0.5 h1:iy+VFUOCP1a+8yFto/drg2CJ5u0yRoB7fZw3DKv/JXA=
|
||||
github.com/spf13/pflag v1.0.5/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
|
||||
github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
|
||||
github.com/stretchr/testify v1.6.1 h1:hDPOHmpOpP40lSULcqw7IrRb/u7w6RpDC9399XyoNd0=
|
||||
github.com/stretchr/testify v1.6.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
|
||||
github.com/youpy/go-riff v0.1.0 h1:vZO/37nI4tIET8tQI0Qn0Y79qQh99aEpponTPiPut7k=
|
||||
github.com/youpy/go-riff v0.1.0/go.mod h1:83nxdDV4Z9RzrTut9losK7ve4hUnxUR8ASSz4BsKXwQ=
|
||||
github.com/youpy/go-wav v0.3.2 h1:NLM8L/7yZ0Bntadw/0h95OyUsen+DQIVf9gay+SUsMU=
|
||||
github.com/youpy/go-wav v0.3.2/go.mod h1:0FCieAXAeSdcxFfwLpRuEo0PFmAoc+8NU34h7TUvk50=
|
||||
github.com/zaf/g711 v0.0.0-20190814101024-76a4a538f52b h1:QqixIpc5WFIqTLxB3Hq8qs0qImAgBdq0p6rq2Qdl634=
|
||||
github.com/zaf/g711 v0.0.0-20190814101024-76a4a538f52b/go.mod h1:T2h1zV50R/q0CVYnsQOQ6L7P4a2ZxH47ixWcMXFGyx8=
|
||||
golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
|
||||
gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
|
||||
gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c h1:dUUwHk2QECo/6vqA44rthZ8ie2QXMNeKRTHCNY2nXvo=
|
||||
gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
|
||||
gotest.tools v2.2.0+incompatible h1:VsBPFP1AI068pPrMxtb/S8Zkgf9xEmTLJjfM+P5UIEo=
|
||||
gotest.tools v2.2.0+incompatible/go.mod h1:DsYFclhRJ6vuDpmuTbkuFWG+y2sxOXAzmJt81HFBacw=
|
||||
118
go-api-examples/streaming-decode-files/main.go
Normal file
118
go-api-examples/streaming-decode-files/main.go
Normal file
@@ -0,0 +1,118 @@
|
||||
package main
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"encoding/binary"
|
||||
sherpa "github.com/k2-fsa/sherpa-onnx-go/sherpa_onnx"
|
||||
flag "github.com/spf13/pflag"
|
||||
"github.com/youpy/go-wav"
|
||||
"os"
|
||||
"strings"
|
||||
|
||||
"log"
|
||||
)
|
||||
|
||||
func main() {
|
||||
log.SetFlags(log.LstdFlags | log.Lmicroseconds)
|
||||
|
||||
config := sherpa.OnlineRecognizerConfig{}
|
||||
config.FeatConfig = sherpa.FeatureConfig{SampleRate: 16000, FeatureDim: 80}
|
||||
|
||||
flag.StringVar(&config.ModelConfig.Encoder, "encoder", "", "Path to the encoder model")
|
||||
flag.StringVar(&config.ModelConfig.Decoder, "decoder", "", "Path to the decoder model")
|
||||
flag.StringVar(&config.ModelConfig.Joiner, "joiner", "", "Path to the joiner 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")
|
||||
flag.StringVar(&config.ModelConfig.ModelType, "model-type", "", "Optional. Used for loading the model in a faster way")
|
||||
flag.StringVar(&config.ModelConfig.Provider, "provider", "cpu", "Provider to use")
|
||||
flag.StringVar(&config.DecodingMethod, "decoding-method", "greedy_search", "Decoding method. Possible values: greedy_search, modified_beam_search")
|
||||
flag.IntVar(&config.MaxActivePaths, "max-active-paths", 4, "Used only when --decoding-method is modified_beam_search")
|
||||
|
||||
flag.Parse()
|
||||
|
||||
if len(flag.Args()) != 1 {
|
||||
log.Fatalf("Please provide one wave file")
|
||||
}
|
||||
|
||||
log.Println("Reading", flag.Arg(0))
|
||||
|
||||
samples, sampleRate := readWave(flag.Arg(0))
|
||||
|
||||
log.Println("Initializing recognizer (may take several seconds)")
|
||||
recognizer := sherpa.NewOnlineRecognizer(&config)
|
||||
log.Println("Recognizer created!")
|
||||
defer sherpa.DeleteOnlineRecognizer(recognizer)
|
||||
|
||||
log.Println("Start decoding!")
|
||||
stream := sherpa.NewOnlineStream(recognizer)
|
||||
defer sherpa.DeleteOnlineStream(stream)
|
||||
|
||||
stream.AcceptWaveform(sampleRate, samples)
|
||||
|
||||
tailPadding := make([]float32, int(float32(sampleRate)*0.3))
|
||||
stream.AcceptWaveform(sampleRate, tailPadding)
|
||||
|
||||
for recognizer.IsReady(stream) {
|
||||
recognizer.Decode(stream)
|
||||
}
|
||||
log.Println("Decoding done!")
|
||||
result := recognizer.GetResult(stream)
|
||||
log.Println(strings.ToLower(result.Text))
|
||||
log.Printf("Wave duration: %v seconds", float32(len(samples))/float32(sampleRate))
|
||||
}
|
||||
|
||||
func readWave(filename string) (samples []float32, sampleRate int) {
|
||||
file, _ := os.Open(filename)
|
||||
defer file.Close()
|
||||
|
||||
reader := wav.NewReader(file)
|
||||
format, err := reader.Format()
|
||||
if err != nil {
|
||||
log.Fatalf("Failed to read wave format")
|
||||
}
|
||||
|
||||
if format.AudioFormat != 1 {
|
||||
log.Fatalf("Support only PCM format. Given: %v\n", format.AudioFormat)
|
||||
}
|
||||
|
||||
if format.NumChannels != 1 {
|
||||
log.Fatalf("Support only 1 channel wave file. Given: %v\n", format.NumChannels)
|
||||
}
|
||||
|
||||
if format.BitsPerSample != 16 {
|
||||
log.Fatalf("Support only 16-bit per sample. Given: %v\n", format.BitsPerSample)
|
||||
}
|
||||
|
||||
reader.Duration() // so that it initializes reader.Size
|
||||
|
||||
buf := make([]byte, reader.Size)
|
||||
n, err := reader.Read(buf)
|
||||
if n != int(reader.Size) {
|
||||
log.Fatalf("Failed to read %v bytes. Returned %v bytes\n", reader.Size, n)
|
||||
}
|
||||
|
||||
samples = samplesInt16ToFloat(buf)
|
||||
sampleRate = int(format.SampleRate)
|
||||
|
||||
return
|
||||
}
|
||||
|
||||
func samplesInt16ToFloat(inSamples []byte) []float32 {
|
||||
numSamples := len(inSamples) / 2
|
||||
outSamples := make([]float32, numSamples)
|
||||
|
||||
for i := 0; i != numSamples; i++ {
|
||||
s := inSamples[i*2 : (i+1)*2]
|
||||
|
||||
var s16 int16
|
||||
buf := bytes.NewReader(s)
|
||||
err := binary.Read(buf, binary.LittleEndian, &s16)
|
||||
if err != nil {
|
||||
log.Fatal("Failed to parse 16-bit sample")
|
||||
}
|
||||
outSamples[i] = float32(s16) / 32768
|
||||
}
|
||||
|
||||
return outSamples
|
||||
}
|
||||
17
go-api-examples/streaming-decode-files/run.sh
Executable file
17
go-api-examples/streaming-decode-files/run.sh
Executable file
@@ -0,0 +1,17 @@
|
||||
#!/usr/bin/env bash
|
||||
|
||||
# Please refer to
|
||||
# https://k2-fsa.github.io/sherpa/onnx/pretrained_models/online-transducer/zipformer-transducer-models.html#csukuangfj-sherpa-onnx-streaming-zipformer-en-2023-06-26-english
|
||||
# to download the model
|
||||
# before you run this script.
|
||||
#
|
||||
# You can switch to a different online model if you need
|
||||
|
||||
./streaming-decode-files \
|
||||
--encoder ./sherpa-onnx-streaming-zipformer-en-2023-06-26/encoder-epoch-99-avg-1-chunk-16-left-128.onnx \
|
||||
--decoder ./sherpa-onnx-streaming-zipformer-en-2023-06-26/decoder-epoch-99-avg-1-chunk-16-left-128.onnx \
|
||||
--joiner ./sherpa-onnx-streaming-zipformer-en-2023-06-26/joiner-epoch-99-avg-1-chunk-16-left-128.onnx \
|
||||
--tokens ./sherpa-onnx-streaming-zipformer-en-2023-06-26/tokens.txt \
|
||||
--model-type zipformer2 \
|
||||
--debug 0 \
|
||||
./sherpa-onnx-streaming-zipformer-en-2023-06-26/test_wavs/0.wav
|
||||
Reference in New Issue
Block a user