Merge pull request #49 from eric2788/voice-silk

change back alpine linux and use silk instead of amr

Author: eric2788

.github/workflows/push-docker.yml

@@ -9,7 +9,7 @@ on:
     tags:
       - '*'
     branches:
-      - master
+      - '*'
 
 jobs:
   test_program:
@@ -67,7 +67,7 @@ jobs:
             docker.io/${{ secrets.DOCKER_USERNAME }}/${{ env.artifact_name }}
           tags: |
             type=edge
-            type=sha,event=branch,branch=master
+            type=sha,event=branch
             type=ref,event=tag
       - name: Build and push
         id: docker_build

.github/workflows/verify.yml

@@ -7,7 +7,7 @@ env:
 on:
   pull_request:
     branches:
-      - master
+      - '*'
 
 jobs:
   test_program:

.gitignore

@@ -5,4 +5,7 @@
 **/logs/**
 session.token
 **/imgtxt/*.png
-**/file/*.yaml
+**/file/*.yaml
+**/*.exe
+**/cache/**
+**/codec/**

Dockerfile

@@ -1,19 +1,20 @@
-FROM golang:1.17 AS builder
+FROM golang:1.17-alpine AS builder
 
 WORKDIR /app
 
 COPY . .
 
 # install timzone data
-# RUN apk add --no-cache tzdata
+RUN apk add --no-cache tzdata
 
 RUN go mod tidy -compat="1.17"
 RUN go mod download
 RUN go build -v -o /go/bin/valbot
 
-FROM linuxserver/ffmpeg
+FROM alpine:latest
+
+RUN apk add --no-cache ffmpeg
 
-RUN ffmpeg -version
 # copy timezone info from builder
 COPY --from=builder /usr/share/zoneinfo /usr/share/zoneinfo
 COPY --from=builder /go/bin/valbot /valbot

aivoice/aivoice.go

@@ -4,33 +4,76 @@ import (
 	"crypto/md5"
 	"encoding/hex"
 	"fmt"
-	ffmpeg_go "github.com/u2takey/ffmpeg-go"
-	"log"
 	"os"
+	"os/exec"
+
+	"github.com/Logiase/MiraiGo-Template/utils"
+	"github.com/eric2788/go-silk/multiplat"
+	ffmpeg_go "github.com/u2takey/ffmpeg-go"
+	silk "github.com/wdvxdr1123/go-silk"
 )
 
+var logger = utils.GetModuleLogger("valbot.aivoice")
+
 const (
 	AudioSamplingRateMP3  = "22050"
 	AudioBitRate          = "12.2k" // in Hz
 	NumberOfAudioChannels = "1"
 	AudioSamplingRateAMR  = "8000"
 )
 
+func WavToSilk(b []byte) (data []byte, err error) {
+	md := md5.Sum(b)
+	tempName := hex.EncodeToString(md[:])
+
+	wav, pcm := tempName+".wav", tempName+".pcm"
+	logger.Infof("writing file: %s", wav)
+	err = os.WriteFile(wav, b, os.ModePerm)
+	if err != nil {
+		return nil, err
+	}
+	defer os.Remove(wav)
+
+	// 2.转换pcm
+	logger.Infof("converting to pcm...")
+	cmd := exec.Command("ffmpeg", "-i", wav, "-f", "s16le", "-ar", "24000", "-ac", "1", pcm)
+	logger.Infof("converted to pcm: %s", pcm)
+	multiplat.HideWindow(cmd)
+	if err = cmd.Run(); err != nil {
+		return nil, err
+	}
+	defer os.Remove(pcm)
+	pcmByte, err := os.ReadFile(pcm)
+	if err != nil {
+		return nil, err
+	}
+	logger.Infof("converting to silk...")
+	defer func ()  {
+		if err == nil {
+			logger.Infof("converted to silk: %s", tempName+".silk")
+		}		
+	}()
+	data, err = silk.EncodePcmBuffToSilk(pcmByte, 24000, 24000, true)
+	return
+}
+
+// WavToAmr Wav To Amr file
+// Deprecated: use WavToSilk instead
 func WavToAmr(b []byte) (data []byte, err error) {
 	hash := md5.Sum(b)
 	name := hex.EncodeToString(hash[:])
 	f, err := os.Create("./" + name + ".wav")
 	if err != nil {
 		return nil, err
 	}
-	log.Println("writing file: ", f.Name())
+	logger.Info("writing file: ", f.Name())
 	_, err = f.Write(b)
 	f.Close()
 	defer removeFile(f.Name())
 	if err != nil {
 		return nil, err
 	}
-	log.Println("converting to amr: ", f.Name())
+	logger.Info("converting to amr: ", f.Name())
 	// Convert to AMR
 	comm := ffmpeg_go.Input(name+".wav").Output(name+".amr", ffmpeg_go.KwArgs{"ar": AudioSamplingRateAMR, "ab": AudioBitRate})
 	//comm := exec.Command("ffmpeg", "-i", "./"+name+".wav", "-ab", AudioBitRate, "-ar", AudioSamplingRateAMR, name+".amr")
@@ -39,7 +82,7 @@ func WavToAmr(b []byte) (data []byte, err error) {
 	}
 	defer removeFile(name + ".amr")
 	data, err = os.ReadFile(name + ".amr")
-	log.Println("successfully converted to amr")
+	logger.Info("successfully converted to amr")
 	return
 }
 

aivoice/aivoice_test.go

@@ -10,6 +10,7 @@ func TestGetGenshinVoice(t *testing.T) {
 	b, err := GetGenshinVoice("別狗叫", "派蒙")
 	if err != nil {
 		if strings.HasPrefix(err.Error(), "http_error") {
+			t.Logf("got error: %v, skip test", err)
 			return
 		}
 		t.Fatal(err)
@@ -21,7 +22,7 @@ func TestGetGenshinVoice(t *testing.T) {
 	if err != nil {
 		t.Fatal(err)
 	}
-	err = os.WriteFile("data/別狗叫.amr", b, os.ModePerm)
+	err = os.WriteFile("data/別狗叫.silk", b, os.ModePerm)
 	if err != nil {
 		t.Fatal(err)
 	}

aivoice/genshin_voice.go

@@ -98,6 +98,6 @@ func GetGenshinVoice(msg, actor string) ([]byte, error) {
 		if err != nil {
 			return nil, err
 		}
-		return WavToAmr(b)
+		return WavToSilk(b)
 	}
 }

cmd/test.go

@@ -0,0 +1,46 @@
+package cmd
+
+import (
+	"fmt"
+
+	"github.com/Mrs4s/MiraiGo/message"
+	"github.com/eric2788/MiraiValBot/modules/command"
+	"github.com/eric2788/MiraiValBot/qq"
+	"github.com/eric2788/common-utils/request"
+)
+
+
+func testSendMp3Voice(args []string, source *command.MessageSource) error {
+	data, err := request.GetBytesByUrl("https://genshin.azurewebsites.net/api/speak?format=mp3&text=测试测试&id=0")
+	if err != nil {
+		return err
+	}
+	voice := &message.VoiceElement{Data: data}
+	return qq.SendGroupMessage(message.NewSendingMessage().Append(voice))
+}
+
+func testSendWavVoice(args []string, source *command.MessageSource) error {
+	data, err := request.GetBytesByUrl("https://genshin.azurewebsites.net/api/speak?format=wav&text=测试测试&id=0")
+	if err != nil {
+		return err
+	}
+	voice := &message.VoiceElement{Data: data}
+	return qq.SendGroupMessage(message.NewSendingMessage().Append(voice))
+}
+
+var testCommands = []command.CmdHandler{
+	testSendMp3Voice,
+	testSendWavVoice,
+}
+
+
+func init(){
+	nodes := make([]command.Node, len(testCommands))
+	for i, handler := range testCommands {
+		name := fmt.Sprintf("%d", i)
+		nodes = append(nodes, command.NewNode([]string{name}, name, true, handler))
+	}
+	var testCommand = command.NewParent([]string{"test", "测试"}, "测试指令", nodes...)
+	testCommand.AdminOnly = true
+	command.AddCommand(testCommand)
+}

cmd/voice.go

@@ -29,9 +29,6 @@ func voiceQQ(args []string, source *command.MessageSource) error {
 		logger.Errorf("tts 转换失败: %v", err)
 		return err
 	}
-
-	logger.Infof("嘗試發送voiceElement: %v", content)
-
 	return qq.SendGroupMessage(message.NewSendingMessage().Append(voiceElement))
 }
 
@@ -45,13 +42,10 @@ func voiceGenshin(args []string, source *command.MessageSource) error {
 	if err != nil {
 		return err
 	}
-
 	voice, err := qq.NewVoiceByBytes(data)
 	if err != nil {
 		return err
 	}
-
-	//voice := &message.GroupVoiceElement{Data: data}
 	return qq.SendGroupMessage(message.NewSendingMessage().Append(voice))
 }
 

go.mod

@@ -8,13 +8,15 @@ require (
 	github.com/bwmarrin/discordgo v0.23.2
 	github.com/corpix/uarand v0.2.0
 	github.com/eric2788/common-utils v0.0.0-20220920151602-fe46ee86cd5a
+	github.com/eric2788/go-silk v0.0.0-20221103160020-81195948ffe2
 	github.com/go-redis/redis/v8 v8.11.4
 	github.com/golang/freetype v0.0.0-20170609003504-e2365dfdc4a0
 	github.com/google/uuid v1.3.0
 	github.com/hqbobo/text2pic v0.0.0-20180823042751-2479e146d720
 	github.com/sirupsen/logrus v1.8.1
 	github.com/stretchr/testify v1.7.1
 	github.com/u2takey/ffmpeg-go v0.4.1
+	github.com/wdvxdr1123/go-silk v0.0.0-20220304095002-f67345df09ea
 	gopkg.in/yaml.v3 v3.0.0-20210107192922-496545a6307b
 )
 
@@ -35,13 +37,14 @@ require (
 	github.com/magiconair/properties v1.8.5 // indirect
 	github.com/maruel/rs v1.0.0 // indirect
 	github.com/mattn/go-colorable v0.1.12 // indirect
-	github.com/mattn/go-isatty v0.0.14 // indirect
+	github.com/mattn/go-isatty v0.0.16 // indirect
 	github.com/mitchellh/mapstructure v1.4.3 // indirect
 	github.com/nfnt/resize v0.0.0-20180221191011-83c6a9932646 // indirect
 	github.com/pelletier/go-toml v1.9.4 // indirect
 	github.com/pierrec/lz4/v4 v4.1.12 // indirect
 	github.com/pkg/errors v0.9.1 // indirect
 	github.com/pmezard/go-difflib v1.0.0 // indirect
+	github.com/remyoudompheng/bigfft v0.0.0-20220927061507-ef77025ab5aa // indirect
 	github.com/rifflock/lfshook v0.0.0-20180920164130-b9218ef580f5 // indirect
 	github.com/skip2/go-qrcode v0.0.0-20200617195104-da1b6568686e // indirect
 	github.com/spf13/afero v1.8.0 // indirect
@@ -64,4 +67,7 @@ require (
 	golang.org/x/text v0.4.0 // indirect
 	gopkg.in/ini.v1 v1.66.2 // indirect
 	gopkg.in/yaml.v2 v2.4.0 // indirect
+	modernc.org/libc v1.21.4 // indirect
+	modernc.org/mathutil v1.5.0 // indirect
+	modernc.org/memory v1.4.0 // indirect
 )