Feat/qwen minimax tts voice clone

.github/workflows/docker-build-branch.yml

@@ -0,0 +1,66 @@
+name: Docker Build & Push
+
+on:
+  push:
+    branches:
+      - 'feat/**'
+      - 'fix/**'
+      - 'test/**'
+      - 'main'
+  workflow_dispatch:
+
+env:
+  REGISTRY: ghcr.io
+  IMAGE_NAME: ${{ github.repository }}
+
+jobs:
+  build:
+    name: Build & push
+    runs-on: ubuntu-latest
+    permissions:
+      packages: write
+      contents: read
+
+    steps:
+      - name: Check out
+        uses: actions/checkout@v4
+
+      - name: Write VERSION
+        run: echo "${GITHUB_REF##*/}-${GITHUB_SHA::7}" > VERSION
+
+      - name: Log in to GHCR
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.REGISTRY }}
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Extract metadata
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
+          tags: |
+            type=ref,event=branch
+            type=sha,prefix=
+            type=raw,value=latest,enable=${{ github.ref == 'refs/heads/main' }}
+
+      - name: Build & push
+        uses: docker/build-push-action@v6
+        with:
+          context: .
+          push: true
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max
+
+      - name: Image digest
+        run: |
+          echo "### Docker Image" >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY
+          echo "${{ steps.meta.outputs.tags }}" >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY

controller/option.go

@@ -259,6 +259,24 @@ func UpdateOption(c *gin.Context) {
 			})
 			return
 		}
+	case "VoiceCloneUnlockRatio":
+		err = ratio_setting.UpdateVoiceCloneUnlockRatioByJSONString(option.Value.(string))
+		if err != nil {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": "音色解锁倍率设置失败: " + err.Error(),
+			})
+			return
+		}
+	case "VideoResolutionRatio":
+		err = ratio_setting.UpdateVideoResolutionRatioByJSONString(option.Value.(string))
+		if err != nil {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": "视频分辨率倍率设置失败: " + err.Error(),
+			})
+			return
+		}
 	case "CreateCacheRatio":
 		err = ratio_setting.UpdateCreateCacheRatioByJSONString(option.Value.(string))
 		if err != nil {

controller/ratio_sync.go

@@ -69,20 +69,24 @@ var pricingSyncFields = []string{
 	"image_ratio",
 	"audio_ratio",
 	"audio_completion_ratio",
+	"voice_clone_unlock_ratio",
+	"video_resolution_ratio",
 	"model_price",
 	billing_setting.BillingModeField,
 	billing_setting.BillingExprField,
 }
 
 var numericPricingSyncFields = map[string]bool{
-	"model_ratio":            true,
-	"completion_ratio":       true,
-	"cache_ratio":            true,
-	"create_cache_ratio":     true,
-	"image_ratio":            true,
-	"audio_ratio":            true,
-	"audio_completion_ratio": true,
-	"model_price":            true,
+	"model_ratio":              true,
+	"completion_ratio":         true,
+	"cache_ratio":              true,
+	"create_cache_ratio":       true,
+	"image_ratio":              true,
+	"audio_ratio":              true,
+	"audio_completion_ratio":   true,
+	"voice_clone_unlock_ratio": true,
+	"video_resolution_ratio":   true,
+	"model_price":              true,
 }
 
 type upstreamResult struct {
@@ -136,6 +140,8 @@ func getLocalPricingSyncData() map[string]any {
 	data["image_ratio"] = ratio_setting.GetImageRatioCopy()
 	data["audio_ratio"] = ratio_setting.GetAudioRatioCopy()
 	data["audio_completion_ratio"] = ratio_setting.GetAudioCompletionRatioCopy()
+	data["voice_clone_unlock_ratio"] = ratio_setting.GetVoiceCloneUnlockRatioCopy()
+	data["video_resolution_ratio"] = ratio_setting.GetVideoResolutionRatioCopy()
 	return data
 }
 

controller/relay.go

@@ -39,6 +39,8 @@ func relayHandler(c *gin.Context, info *relaycommon.RelayInfo) *types.NewAPIErro
 		err = relay.ImageHelper(c, info)
 	case relayconstant.RelayModeAudioSpeech:
 		fallthrough
+	case relayconstant.RelayModeAudioVoiceClone:
+		fallthrough
 	case relayconstant.RelayModeAudioTranslation:
 		fallthrough
 	case relayconstant.RelayModeAudioTranscription:

dto/audio.go

@@ -30,6 +30,17 @@ type AudioRequest struct {
 	//Stream                  json.RawMessage `json:"stream,omitempty"`
 }
 
+type AudioVoiceCloneRequest struct {
+	Model    string          `json:"model"`
+	Input    json.RawMessage `json:"input,omitempty"`
+	Text     string          `json:"text,omitempty"`
+	VoiceID  string          `json:"voice_id,omitempty"`
+	FileID   int64           `json:"file_id,omitempty"`
+	AudioURL string          `json:"audio_url,omitempty"`
+	Metadata json.RawMessage `json:"metadata,omitempty"`
+	rawText  string
+}
+
 func (r *AudioRequest) GetTokenCountMeta() *types.TokenCountMeta {
 	meta := &types.TokenCountMeta{
 		CombineText: r.Input,
@@ -41,6 +52,32 @@ func (r *AudioRequest) GetTokenCountMeta() *types.TokenCountMeta {
 	return meta
 }
 
+func (r *AudioVoiceCloneRequest) GetTokenCountMeta() *types.TokenCountMeta {
+	text := r.Text
+	if text == "" && len(r.Input) > 0 {
+		var input struct {
+			Text string `json:"text"`
+		}
+		if json.Unmarshal(r.Input, &input) == nil {
+			text = input.Text
+		}
+	}
+	return &types.TokenCountMeta{
+		CombineText: text,
+		TokenType:   types.TokenTypeTextNumber,
+	}
+}
+
+func (r *AudioVoiceCloneRequest) IsStream(c *gin.Context) bool {
+	return false
+}
+
+func (r *AudioVoiceCloneRequest) SetModelName(modelName string) {
+	if modelName != "" {
+		r.Model = modelName
+	}
+}
+
 func (r *AudioRequest) IsStream(c *gin.Context) bool {
 	return r.StreamFormat == "sse"
 }

model/option.go

@@ -148,6 +148,8 @@ func InitOptionMap() {
 	common.OptionMap["ImageRatio"] = ratio_setting.ImageRatio2JSONString()
 	common.OptionMap["AudioRatio"] = ratio_setting.AudioRatio2JSONString()
 	common.OptionMap["AudioCompletionRatio"] = ratio_setting.AudioCompletionRatio2JSONString()
+	common.OptionMap["VoiceCloneUnlockRatio"] = ratio_setting.VoiceCloneUnlockRatio2JSONString()
+	common.OptionMap["VideoResolutionRatio"] = ratio_setting.VideoResolutionRatio2JSONString()
 	common.OptionMap["TopUpLink"] = common.TopUpLink
 	//common.OptionMap["ChatLink"] = common.ChatLink
 	//common.OptionMap["ChatLink2"] = common.ChatLink2
@@ -540,6 +542,10 @@ func updateOptionMap(key string, value string) (err error) {
 		err = ratio_setting.UpdateAudioRatioByJSONString(value)
 	case "AudioCompletionRatio":
 		err = ratio_setting.UpdateAudioCompletionRatioByJSONString(value)
+	case "VoiceCloneUnlockRatio":
+		err = ratio_setting.UpdateVoiceCloneUnlockRatioByJSONString(value)
+	case "VideoResolutionRatio":
+		err = ratio_setting.UpdateVideoResolutionRatioByJSONString(value)
 	case "TopUpLink":
 		common.TopUpLink = value
 	//case "ChatLink":

model/pricing.go

@@ -31,6 +31,7 @@ type Pricing struct {
 	ImageRatio             *float64                `json:"image_ratio,omitempty"`
 	AudioRatio             *float64                `json:"audio_ratio,omitempty"`
 	AudioCompletionRatio   *float64                `json:"audio_completion_ratio,omitempty"`
+	VoiceCloneUnlockRatio  *float64                `json:"voice_clone_unlock_ratio,omitempty"`
 	EnableGroup            []string                `json:"enable_groups"`
 	SupportedEndpointTypes []constant.EndpointType `json:"supported_endpoint_types"`
 	BillingMode            string                  `json:"billing_mode,omitempty"`
@@ -331,6 +332,9 @@ func updatePricing() {
 			audioCompletionRatio := ratio_setting.GetAudioCompletionRatio(model)
 			pricing.AudioCompletionRatio = &audioCompletionRatio
 		}
+		if voiceCloneUnlockRatio, ok := ratio_setting.GetVoiceCloneUnlockRatio(model); ok {
+			pricing.VoiceCloneUnlockRatio = &voiceCloneUnlockRatio
+		}
 		if billingMode := billing_setting.GetBillingMode(model); billingMode == "tiered_expr" {
 			if expr, ok := billing_setting.GetBillingExpr(model); ok && strings.TrimSpace(expr) != "" {
 				pricing.BillingMode = billingMode

relay/audio_handler.go

@@ -19,16 +19,12 @@ func AudioHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *type
 	info.InitChannelMeta(c)
 
 	audioReq, ok := info.Request.(*dto.AudioRequest)
-	if !ok {
+	voiceCloneReq, cloneOK := info.Request.(*dto.AudioVoiceCloneRequest)
+	if !ok && !cloneOK {
 		return types.NewError(errors.New("invalid request type"), types.ErrorCodeInvalidRequest, types.ErrOptionWithSkipRetry())
 	}
 
-	request, err := common.DeepCopy(audioReq)
-	if err != nil {
-		return types.NewError(fmt.Errorf("failed to copy request to AudioRequest: %w", err), types.ErrorCodeInvalidRequest, types.ErrOptionWithSkipRetry())
-	}
-
-	err = helper.ModelMappedHelper(c, info, request)
+	err := helper.ModelMappedHelper(c, info, info.Request)
 	if err != nil {
 		return types.NewError(err, types.ErrorCodeChannelModelMappedError, types.ErrOptionWithSkipRetry())
 	}
@@ -39,7 +35,19 @@ func AudioHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *type
 	}
 	adaptor.Init(info)
 
-	ioReader, err := adaptor.ConvertAudioRequest(c, info, *request)
+	request := dto.AudioRequest{}
+	if ok {
+		copied, copyErr := common.DeepCopy(audioReq)
+		if copyErr != nil {
+			return types.NewError(fmt.Errorf("failed to copy request to AudioRequest: %w", copyErr), types.ErrorCodeInvalidRequest, types.ErrOptionWithSkipRetry())
+		}
+		request = *copied
+	} else if cloneOK {
+		request.Model = voiceCloneReq.Model
+		request.Input = voiceCloneReq.Text
+	}
+
+	ioReader, err := adaptor.ConvertAudioRequest(c, info, request)
 	if err != nil {
 		return types.NewError(err, types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
 	}

relay/channel/ali/adaptor.go

@@ -119,6 +119,14 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 			} else {
 				fullRequestURL = fmt.Sprintf("%s/api/v1/services/aigc/multimodal-generation/generation", info.ChannelBaseUrl)
 			}
+		case constant.RelayModeAudioSpeech:
+			fullRequestURL = fmt.Sprintf("%s/api/v1/services/aigc/multimodal-generation/generation", info.ChannelBaseUrl)
+		case constant.RelayModeAudioVoiceClone:
+			if isAliMiniMaxSpeechModel(info.UpstreamModelName) || isAliMiniMaxSpeechModel(info.OriginModelName) {
+				fullRequestURL = fmt.Sprintf("%s/api/v1/services/aigc/multimodal-generation/generation", info.ChannelBaseUrl)
+			} else {
+				fullRequestURL = fmt.Sprintf("%s/api/v1/services/audio/tts/customization", info.ChannelBaseUrl)
+			}
 		case constant.RelayModeCompletions:
 			fullRequestURL = fmt.Sprintf("%s/compatible-mode/v1/completions", info.ChannelBaseUrl)
 		default:
@@ -151,6 +159,9 @@ func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *rel
 		}
 		req.Set("Content-Type", "application/json")
 	}
+	if info.RelayMode == constant.RelayModeAudioSpeech || info.RelayMode == constant.RelayModeAudioVoiceClone {
+		req.Set("Content-Type", "application/json")
+	}
 	return nil
 }
 
@@ -226,8 +237,14 @@ func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.Rela
 }
 
 func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.AudioRequest) (io.Reader, error) {
-	//TODO implement me
-	return nil, errors.New("not implemented")
+	switch info.RelayMode {
+	case constant.RelayModeAudioSpeech:
+		return convertOpenAIToAliTTS(c, info, request)
+	case constant.RelayModeAudioVoiceClone:
+		return convertAliVoiceClone(c, info, request)
+	default:
+		return nil, errors.New("unsupported audio relay mode")
+	}
 }
 
 func (a *Adaptor) ConvertOpenAIResponsesRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.OpenAIResponsesRequest) (any, error) {
@@ -256,6 +273,10 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 			err, usage = aliImageHandler(a, c, resp, info)
 		case constant.RelayModeRerank:
 			err, usage = RerankHandler(c, resp, info)
+		case constant.RelayModeAudioSpeech:
+			err, usage = aliTTSHandler(c, resp, info)
+		case constant.RelayModeAudioVoiceClone:
+			err, usage = aliVoiceCloneHandler(c, resp, info)
 		default:
 			adaptor := openai.Adaptor{}
 			usage, err = adaptor.DoResponse(c, resp, info)

relay/channel/ali/constants.go

@@ -1,12 +1,25 @@
 package ali
 
+// https://help.aliyun.com/zh/model-studio/voice-clone-design-http-api
+// https://help.aliyun.com/zh/model-studio/voice-cloning-user-guide
+
 var ModelList = []string{
+	// Chat models
 	"qwen-turbo",
 	"qwen-plus",
 	"qwen-max",
 	"qwen-max-longcontext",
 	"qwq-32b",
 	"qwen3-235b-a22b",
+	// TTS models - 语音合成
+	"qwen-tts",
+	"qwen-tts-latest",
+	"qwen3-tts-flash",
+	"qwen3-tts-vc-realtime-2026-01-15",
+	// Voice clone models - 语音克隆
+	"qwen-voice-clone",
+	"qwen-voice-enrollment",
+	// Embedding and rerank models
 	"text-embedding-v1",
 	"gte-rerank-v2",
 }

relay/channel/ali/dto.go

@@ -73,6 +73,7 @@ type AliUsage struct {
 	OutputTokens int `json:"output_tokens"`
 	TotalTokens  int `json:"total_tokens"`
 	ImageCount   int `json:"image_count,omitempty"`
+	Count        int `json:"count,omitempty"`
 }
 
 type TaskResult struct {