Merge pull request #31 from mutablelogic/v1

Reorganized bindings and service

Author: djthorpe

Makefile

@@ -25,7 +25,7 @@ server: mkdir go-tidy libwhisper libggml
 	 ${GO} build -o ${BUILD_DIR}/whisper-server ./cmd/server
 
 # Make cli
-cli: mkdir go-tidy
+cli: mkdir go-tidy libwhisper libggml
 	@echo "Building whisper-cli"
 	@CGO_CFLAGS="-I${ROOT_PATH}/third_party/whisper.cpp/include -I${ROOT_PATH}/third_party/whisper.cpp/ggml/include" \
 	 CGO_LDFLAGS="-L${ROOT_PATH}/third_party/whisper.cpp" \

go.mod

@@ -7,7 +7,7 @@ require (
 	github.com/djthorpe/go-tablewriter v0.0.7
 	github.com/go-audio/wav v1.1.0
 	github.com/mutablelogic/go-client v1.0.8
-	github.com/mutablelogic/go-server v1.4.10
+	github.com/mutablelogic/go-server v1.4.11
 	github.com/stretchr/testify v1.9.0
 )
 

go.sum

@@ -22,15 +22,17 @@ github.com/mattn/go-runewidth v0.0.15 h1:UNAjwbU9l54TA3KzvqLGxwWjHmMgBUVhBiTjelZ
 github.com/mattn/go-runewidth v0.0.15/go.mod h1:Jdepj2loyihRzMpdS35Xk/zdY8IAYHsh153qUoGf23w=
 github.com/mutablelogic/go-client v1.0.8 h1:A3QtP0wdf+W3dE5k7dobwGYqqn4ZpIqRFu+h9vPoy7Y=
 github.com/mutablelogic/go-client v1.0.8/go.mod h1:aP9ecBd4R/acJEJSyp81U3mey9W3AHQV/G1XzfcrLx0=
-github.com/mutablelogic/go-server v1.4.10 h1:iItbUmynlfsIpRHjZXXGJHnYCHFmRgokkn/DIbInR6k=
-github.com/mutablelogic/go-server v1.4.10/go.mod h1:q5zurZQ/DcdGnssptGXxtady5o/ZdhL/5VPnKKED1J0=
+github.com/mutablelogic/go-server v1.4.11 h1:feI9IyuK6pv7Gi7fbExfU51uDRMfQo0U9wo0vWN2wf8=
+github.com/mutablelogic/go-server v1.4.11/go.mod h1:9nenPAohKu8bFoRgwHJh+3s8h0kLFjUAb8KZvT1TQNU=
 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/rivo/uniseg v0.2.0/go.mod h1:J6wj4VEh+S6ZtnVlnTBMWIodfgj8LQOQFoIToxlJtxc=
 github.com/rivo/uniseg v0.4.7 h1:WUdvkW8uEhrYfLC4ZzdpI2ztxP1I582+49Oc5Mq64VQ=
 github.com/rivo/uniseg v0.4.7/go.mod h1:FN3SvrM+Zdj16jyLfmOkMNblXMcoc8DfTHruCPUcx88=
 github.com/stretchr/testify v1.9.0 h1:HtqpIVDClZ4nwg75+f6Lvsy/wHu+3BoSGCbBAcpTsTg=
 github.com/stretchr/testify v1.9.0/go.mod h1:r2ic/lqez/lEtzL7wO/rwa5dbSLXVDPFyf8C91i36aY=
+golang.org/x/exp v0.0.0-20240506185415-9bf2ced13842 h1:vr/HnozRka3pE4EsMEg1lgkXJkTFJCVUX+S/ZT6wYzM=
+golang.org/x/exp v0.0.0-20240506185415-9bf2ced13842/go.mod h1:XtvwrStGgqGPLc4cjQfWqZHG1YFdYs6swckp8vpsjnc=
 golang.org/x/sys v0.21.0 h1:rF+pYz3DAGSQAxAu1CbC7catZg4ebC4UIeIhKxBZvws=
 golang.org/x/sys v0.21.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
 golang.org/x/term v0.21.0 h1:WVXCp+/EBEHOj53Rvu+7KiT/iElMrO8ACK16SMZ3jaA=

pkg/whisper/context/context.go

@@ -0,0 +1,97 @@
+package pool
+
+import (
+
+	// Packages
+
+	"path/filepath"
+
+	model "github.com/mutablelogic/go-whisper/pkg/whisper/model"
+	"github.com/mutablelogic/go-whisper/sys/whisper"
+
+	// Namespace imports
+	. "github.com/djthorpe/go-errors"
+)
+
+//////////////////////////////////////////////////////////////////////////////
+// TYPES
+
+// Context is used for running the transcription or translation
+type Context struct {
+	Model *model.Model
+
+	whisper *whisper.Context
+	params  whisper.FullParams
+}
+
+//////////////////////////////////////////////////////////////////////////////
+// LIFECYCLE
+
+// Create a new context object
+func New(path string, model *model.Model) (*Context, error) {
+	ctx := new(Context)
+
+	// Init the context
+	if err := ctx.Init(path, model); err != nil {
+		return nil, err
+	}
+
+	// Init the transcription with default parameters
+	ctx.params = whisper.DefaultFullParams(whisper.SAMPLING_GREEDY)
+
+	// Return success
+	return ctx, nil
+}
+
+// Init the context
+func (m *Context) Init(path string, model *model.Model) error {
+	// Check parameters
+	if model == nil {
+		return ErrBadParameter
+	}
+
+	// Get a context
+	ctx := whisper.Whisper_init_from_file_with_params(filepath.Join(path, model.Path), whisper.DefaultContextParams())
+	if ctx == nil {
+		return ErrInternalAppError.With("whisper_init_from_file_with_params")
+	}
+
+	// Set resources
+	m.whisper = ctx
+	m.Model = model
+
+	// Return success
+	return nil
+}
+
+// Close the context and release all resources
+func (m *Context) Close() error {
+	var result error
+
+	// Do nothing if nil
+	if m == nil {
+		return nil
+	}
+
+	// Release resources
+	if m.whisper != nil {
+		whisper.Whisper_free(m.whisper)
+	}
+	m.whisper = nil
+	m.Model = nil
+
+	// Return any errors
+	return result
+}
+
+//////////////////////////////////////////////////////////////////////////////
+// PUBLIC METHODS
+
+// Transcribe samples. The samples should be 16KHz float32 samples in
+// a single channel.
+// TODO: We need a low-latency streaming version of this function.
+// TODO: We need a callback for segment progress.
+func (ctx *Context) Transcribe(samples []float32) error {
+	// Perform the transcription
+	return whisper.Whisper_full(ctx.whisper, ctx.params, samples)
+}

pkg/whisper/context.go renamed to pkg/whisper/context/context.go_old

@@ -0,0 +1,27 @@
+package model
+
+import (
+	"encoding/json"
+)
+
+//////////////////////////////////////////////////////////////////////////////
+// TYPES
+
+type Model struct {
+	Id      string `json:"id" writer:",width:28,wrap"`
+	Object  string `json:"object,omitempty" writer:"-"`
+	Path    string `json:"path,omitempty" writer:",width:40,wrap"`
+	Created int64  `json:"created,omitempty"`
+	OwnedBy string `json:"owned_by,omitempty"`
+}
+
+//////////////////////////////////////////////////////////////////////////////
+// STRINGIFY
+
+func (m *Model) String() string {
+	data, err := json.MarshalIndent(m, "", "  ")
+	if err != nil {
+		return err.Error()
+	}
+	return string(data)
+}