Support for flac and opus for batch transcription

.github/workflows/golangci-lint.yaml

@@ -12,7 +12,7 @@ jobs:
     steps:
       - uses: actions/setup-go@v3
         with:
-          go-version: 1.17
+          go-version: 1.18
       - uses: actions/checkout@v3
       - name: golangci-lint
         uses: golangci/golangci-lint-action@v3

.github/workflows/release.yml

@@ -12,7 +12,7 @@ jobs:
     steps:
       - uses: actions/setup-go@v3
         with:
-          go-version: 1.17
+          go-version: 1.18
       - uses: actions/checkout@v3
       - name: golangci-lint
         uses: golangci/golangci-lint-action@v3
@@ -24,7 +24,7 @@ jobs:
       - name: Install Go
         uses: actions/setup-go@v3
         with:
-          go-version: 1.17
+          go-version: 1.18
       - uses: actions/checkout@v3
       - name: build
         run: make build
@@ -41,7 +41,7 @@ jobs:
           fetch-depth: 0
       - uses: actions/setup-go@v3
         with:
-          go-version: 1.17
+          go-version: 1.18
       - uses: docker/login-action@v2
         with:
           username: ${{ secrets.DOCKERHUB_USERNAME }}

.github/workflows/test.yaml

@@ -12,7 +12,7 @@ jobs:
       - name: Install Go
         uses: actions/setup-go@v3
         with:
-          go-version: 1.17
+          go-version: 1.18
       - uses: actions/checkout@v3
       - name: build
         run: make build

cmd/common.go

@@ -233,7 +233,7 @@ func readAudioCorpus(filename string) ([]AudioCorpusItem, error) {
 		return nil, err
 	}
 	ac := make([]AudioCorpusItem, 0)
-	if strings.HasSuffix(filename, "wav") {
+	if strings.HasSuffix(filename, "wav") || strings.HasSuffix(filename, "opus") || strings.HasSuffix(filename, "flac") {
 		return []AudioCorpusItem{{Audio: filename}}, nil
 	}
 	jd := json.NewDecoder(f)
@@ -325,15 +325,19 @@ func transcribeWithBatchAPI(ctx context.Context, appID string, corpusPath string
 			audioFilePath = corpusPath
 		}
 
-		err = readAudio(audioFilePath, aci, func(buffer audio.IntBuffer, n int) error {
-			buffer16 := make([]uint16, len(buffer.Data))
-			for i, x := range buffer.Data {
-				buffer16[i] = uint16(x)
-			}
-			buf := new(bytes.Buffer)
-			err = binary.Write(buf, binary.LittleEndian, buffer16)
-			if err != nil {
-				return fmt.Errorf("binary.Write: %v", err)
+		f, err := os.Open(audioFilePath)
+		if err != nil {
+			barClearOnError(bar)
+			return nil, err
+		}
+		buf := make([]byte, 65536)
+		for {
+			n, err := f.Read(buf)
+			if err == io.EOF {
+				break
+			} else if err != nil {
+				barClearOnError(bar)
+				return nil, err
 			}
 
 			err = paStream.Send(&sluv1.ProcessAudioRequest{
@@ -343,16 +347,14 @@ func transcribeWithBatchAPI(ctx context.Context, appID string, corpusPath string
 					Channels:        1,
 					SampleRateHertz: 16000,
 				},
-				Source: &sluv1.ProcessAudioRequest_Audio{Audio: buf.Bytes()},
+				Source: &sluv1.ProcessAudioRequest_Audio{Audio: buf[:n]},
 			})
+
 			if err != nil {
-				return fmt.Errorf("sending %d process audio request failed: %w", buf.Len(), err)
+				barClearOnError(bar)
+				return nil, err
 			}
-			return nil
-		})
-		if err != nil {
-			barClearOnError(bar)
-			return nil, err
+
 		}
 
 		err = bar.Add(1)
@@ -362,6 +364,10 @@ func transcribeWithBatchAPI(ctx context.Context, appID string, corpusPath string
 		}
 
 		paResp, err := paStream.CloseAndRecv()
+		if err != nil {
+			barClearOnError(bar)
+			return nil, err
+		}
 		bID := paResp.GetOperation().GetId()
 		pending[bID] = aci
 	}
@@ -375,6 +381,7 @@ func transcribeWithBatchAPI(ctx context.Context, appID string, corpusPath string
 	var results []AudioCorpusItem
 
 	bar = getBar("Transcribing", "utt", inputSize)
+	isDone := false
 	for {
 		for bID, aci := range pending {
 			status, err := client.QueryStatus(ctx, &sluv1.QueryStatusRequest{Id: bID})
@@ -386,6 +393,13 @@ func transcribeWithBatchAPI(ctx context.Context, appID string, corpusPath string
 			case sluv1.Operation_STATUS_DONE:
 				trs := status.GetOperation().GetTranscripts()
 				words := make([]string, len(trs))
+				if !isDone && len(trs) == 0 {
+					// Results might not be available immediately after done state is reached, so if we do
+					// not have any, let's wait for a bit.
+					isDone = true
+					time.Sleep(2 * time.Second)
+					continue
+				}
 				for i, tr := range trs {
 					words[i] = tr.Word
 				}

go.mod

@@ -1,48 +1,45 @@
 module github.com/speechly/cli
 
-go 1.17
+go 1.18
 
 require (
 	github.com/agnivade/levenshtein v1.1.1
 	github.com/go-audio/audio v1.0.0
 	github.com/go-audio/wav v1.1.0
-	github.com/mattn/go-isatty v0.0.16
+	github.com/mattn/go-isatty v0.0.17
 	github.com/mitchellh/go-homedir v1.1.0
-	github.com/schollz/progressbar/v3 v3.11.0
-	github.com/speechly/api/go v0.0.0-20220920060221-2531f4783d08
-	github.com/speechly/nwalgo v0.0.0-20221109101309-d1a337619dd3
-	github.com/spf13/cobra v1.3.0
-	github.com/spf13/viper v1.10.0
-	golang.org/x/text v0.4.0
-	google.golang.org/grpc v1.42.0
-	google.golang.org/protobuf v1.27.1
+	github.com/schollz/progressbar/v3 v3.13.0
+	github.com/speechly/api/go v0.0.0-20230221135950-6d68efe6ac91
+	github.com/speechly/nwalgo v0.0.0-20221109110948-f6606115e74b
+	github.com/spf13/cobra v1.6.1
+	github.com/spf13/pflag v1.0.5
+	github.com/spf13/viper v1.15.0
+	golang.org/x/text v0.7.0
+	google.golang.org/grpc v1.53.0
+	google.golang.org/protobuf v1.28.1
 )
 
 require (
-	github.com/cpuguy83/go-md2man/v2 v2.0.1 // indirect
-	github.com/fsnotify/fsnotify v1.5.1 // indirect
+	github.com/fsnotify/fsnotify v1.6.0 // indirect
 	github.com/go-audio/riff v1.0.0 // indirect
 	github.com/golang/protobuf v1.5.2 // indirect
 	github.com/hashicorp/hcl v1.0.0 // indirect
-	github.com/inconshreveable/mousetrap v1.0.0 // indirect
-	github.com/kr/text v0.2.0 // indirect
-	github.com/magiconair/properties v1.8.5 // indirect
-	github.com/mattn/go-runewidth v0.0.13 // indirect
+	github.com/inconshreveable/mousetrap v1.1.0 // indirect
+	github.com/magiconair/properties v1.8.7 // indirect
+	github.com/mattn/go-runewidth v0.0.14 // indirect
 	github.com/mitchellh/colorstring v0.0.0-20190213212951-d06e56a500db // indirect
-	github.com/mitchellh/mapstructure v1.4.3 // indirect
-	github.com/pelletier/go-toml v1.9.4 // indirect
-	github.com/rivo/uniseg v0.2.0 // indirect
-	github.com/russross/blackfriday/v2 v2.1.0 // indirect
-	github.com/spf13/afero v1.6.0 // indirect
-	github.com/spf13/cast v1.4.1 // indirect
+	github.com/mitchellh/mapstructure v1.5.0 // indirect
+	github.com/pelletier/go-toml/v2 v2.0.6 // indirect
+	github.com/rivo/uniseg v0.4.4 // indirect
+	github.com/spf13/afero v1.9.4 // indirect
+	github.com/spf13/cast v1.5.0 // indirect
 	github.com/spf13/jwalterweatherman v1.1.0 // indirect
-	github.com/spf13/pflag v1.0.5 // indirect
-	github.com/subosito/gotenv v1.2.0 // indirect
-	golang.org/x/net v0.0.0-20220722155237-a158d28d115b // indirect
-	golang.org/x/sys v0.0.0-20220829200755-d48e67d00261 // indirect
-	golang.org/x/term v0.0.0-20220722155259-a9ba230a4035 // indirect
-	google.golang.org/genproto v0.0.0-20211208223120-3a66f561d7aa // indirect
+	github.com/subosito/gotenv v1.4.2 // indirect
+	golang.org/x/net v0.7.0 // indirect
+	golang.org/x/sys v0.5.0 // indirect
+	golang.org/x/term v0.5.0 // indirect
+	google.golang.org/genproto v0.0.0-20230223222841-637eb2293923 // indirect
 	gopkg.in/check.v1 v1.0.0-20201130134442-10cb98267c6c // indirect
-	gopkg.in/ini.v1 v1.66.2 // indirect
-	gopkg.in/yaml.v2 v2.4.0 // indirect
+	gopkg.in/ini.v1 v1.67.0 // indirect
+	gopkg.in/yaml.v3 v3.0.1 // indirect
 )