feat: support ali tts

controller/relay.go

@@ -29,7 +29,7 @@ func relayHelper(c *gin.Context, relayMode int) *model.ErrorWithStatusCode {
 	case relaymode.ImagesGenerations:
 		err = controller.RelayImageHelper(c, relayMode)
 	case relaymode.AudioSpeech:
-		fallthrough
+		err = controller.RelayAudioSpeechHelper(c)
 	case relaymode.AudioTranslation:
 		fallthrough
 	case relaymode.AudioTranscription:

relay/adaptor/aiproxy/adaptor.go

@@ -45,6 +45,13 @@ func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error)
 	return request, nil
 }
 
+func (a *Adaptor) ConvertTextToSpeechRequest(request *model.TextToSpeechRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
 	return adaptor.DoRequestHelper(a, c, meta, requestBody)
 }

relay/adaptor/ali/adaptor.go

@@ -1,9 +1,11 @@
 package ali
 
 import (
+	"encoding/json"
 	"errors"
 	"fmt"
 	"github.com/gin-gonic/gin"
+	"github.com/gorilla/websocket"
 	"github.com/songquanpeng/one-api/relay/adaptor"
 	"github.com/songquanpeng/one-api/relay/meta"
 	"github.com/songquanpeng/one-api/relay/model"
@@ -76,7 +78,19 @@ func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error)
 	return aliRequest, nil
 }
 
+func (a *Adaptor) ConvertTextToSpeechRequest(request *model.TextToSpeechRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+
+	aliRequest := ConvertTextToSpeechRequest(*request)
+	return aliRequest, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
+	if meta.Mode == relaymode.AudioSpeech {
+		return a.DoWSSRequest(c, meta, requestBody)
+	}
 	return adaptor.DoRequestHelper(a, c, meta, requestBody)
 }
 
@@ -89,6 +103,8 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *meta.Met
 			err, usage = EmbeddingHandler(c, resp)
 		case relaymode.ImagesGenerations:
 			err, usage = ImageHandler(c, resp)
+		case relaymode.AudioSpeech:
+			err, usage = AudioSpeechHandler(c, resp)
 		default:
 			err, usage = Handler(c, resp)
 		}
@@ -103,3 +119,74 @@ func (a *Adaptor) GetModelList() []string {
 func (a *Adaptor) GetChannelName() string {
 	return "ali"
 }
+
+func (a *Adaptor) DoWSSRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
+	baseURL := "wss://dashscope.aliyuncs.com/api-ws/v1/inference"
+	var usage Usage
+	// Create an empty http.Response object
+	response := &http.Response{
+		StatusCode: http.StatusInternalServerError,
+		Body:       io.NopCloser(nil),
+	}
+
+	conn, _, err := websocket.DefaultDialer.Dial(baseURL, http.Header{"Authorization": {"Bearer " + meta.APIKey}})
+	if err != nil {
+		return response, errors.New("ali_wss_conn_failed")
+	}
+	defer conn.Close()
+
+	var requestBodyBytes []byte
+	requestBodyBytes, err = io.ReadAll(requestBody)
+	if err != nil {
+		return response, errors.New("ali_failed_to_read_request_body")
+	}
+
+	// Convert JSON strings to map[string]interface{}
+	var requestBodyMap map[string]interface{}
+	err = json.Unmarshal(requestBodyBytes, &requestBodyMap)
+	if err != nil {
+		return response, errors.New("ali_failed_to_parse_request_body")
+	}
+
+	if err := conn.WriteJSON(requestBodyMap); err != nil {
+		return response, errors.New("ali_wss_write_msg_failed")
+	}
+
+	const chunkSize = 1024
+
+	for {
+		messageType, audioData, err := conn.ReadMessage()
+		if err != nil {
+			if err == io.EOF {
+				break
+			}
+			return response, errors.New("ali_wss_read_msg_failed")
+		}
+
+		var msg WSSMessage
+		switch messageType {
+		case websocket.TextMessage:
+			err = json.Unmarshal(audioData, &msg)
+			if msg.Header.Event == "task-finished" {
+				response.StatusCode = http.StatusOK
+				usage.TotalTokens = msg.Payload.Usage.Characters
+				return response, nil
+			}
+		case websocket.BinaryMessage:
+			for i := 0; i < len(audioData); i += chunkSize {
+				end := i + chunkSize
+				if end > len(audioData) {
+					end = len(audioData)
+				}
+				chunk := audioData[i:end]
+
+				_, writeErr := c.Writer.Write(chunk)
+				if writeErr != nil {
+					return response, errors.New("wss_write_chunk_failed")
+				}
+			}
+		}
+	}
+
+	return response, nil
+}

relay/adaptor/ali/audio-speech.go

@@ -0,0 +1,21 @@
+package ali
+
+import (
+	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/relay/adaptor/openai"
+	"github.com/songquanpeng/one-api/relay/model"
+	"net/http"
+)
+
+func AudioSpeechHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
+	for k, v := range resp.Header {
+		c.Writer.Header().Set(k, v[0])
+	}
+	c.Writer.WriteHeader(resp.StatusCode)
+
+	err := resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	return nil, nil
+}

relay/adaptor/ali/constants.go

@@ -4,4 +4,48 @@ var ModelList = []string{
 	"qwen-turbo", "qwen-plus", "qwen-max", "qwen-max-longcontext",
 	"text-embedding-v1",
 	"ali-stable-diffusion-xl", "ali-stable-diffusion-v1.5", "wanx-v1",
+
+	"sambert-zhichu-v1",
+	"sambert-zhiwei-v1",
+	"sambert-zhixiang-v1",
+	"sambert-zhide-v1",
+	"sambert-zhijia-v1",
+	"sambert-zhinan-v1",
+	"sambert-zhiqi-v1",
+	"sambert-zhiqian-v1",
+	"sambert-zhiru-v1",
+	"sambert-zhimiao-emo-v1",
+	"sambert-zhida-v1",
+	"sambert-zhifei-v1",
+	"sambert-zhigui-v1",
+	"sambert-zhihao-v1",
+	"sambert-zhijing-v1",
+	"sambert-zhilun-v1",
+	"sambert-zhimao-v1",
+	"sambert-zhiming-v1",
+	"sambert-zhimo-v1",
+	"sambert-zhina-v1",
+	"sambert-zhishu-v1",
+	"sambert-zhishuo-v1",
+	"sambert-zhistella-v1",
+	"sambert-zhiting-v1",
+	"sambert-zhixiao-v1",
+	"sambert-zhiya-v1",
+	"sambert-zhiye-v1",
+	"sambert-zhiying-v1",
+	"sambert-zhiyuan-v1",
+	"sambert-zhiyue-v1",
+	"sambert-camila-v1",
+	"sambert-perla-v1",
+	"sambert-indah-v1",
+	"sambert-clara-v1",
+	"sambert-hanna-v1",
+	"sambert-beth-v1",
+	"sambert-betty-v1",
+	"sambert-cally-v1",
+	"sambert-cindy-v1",
+	"sambert-eva-v1",
+	"sambert-donna-v1",
+	"sambert-brian-v1",
+	"sambert-waan-v1",
 }

relay/adaptor/ali/main.go

@@ -4,6 +4,7 @@ import (
 	"bufio"
 	"encoding/json"
 	"github.com/gin-gonic/gin"
+	"github.com/google/uuid"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/helper"
 	"github.com/songquanpeng/one-api/common/logger"
@@ -77,6 +78,37 @@ func ConvertImageRequest(request model.ImageRequest) *ImageRequest {
 	return &imageRequest
 }
 
+func ConvertTextToSpeechRequest(request model.TextToSpeechRequest) *WSSMessage {
+	var ttsRequest WSSMessage
+	ttsRequest.Header.Action = "run-task"
+	ttsRequest.Header.Streaming = "out"
+	ttsRequest.Header.TaskID = uuid.New().String()
+	ttsRequest.Payload.Function = "SpeechSynthesizer"
+	ttsRequest.Payload.Input.Text = request.Input
+	ttsRequest.Payload.Model = request.Model
+	ttsRequest.Payload.Parameters.Format = "wav"
+	//ttsRequest.Payload.Parameters.SampleRate = 48000
+	ttsRequest.Payload.Parameters.Rate = 1.0
+	ttsRequest.Payload.Task = "tts"
+	ttsRequest.Payload.TaskGroup = "audio"
+
+	format := map[string]bool{
+		"pcm": true,
+		"wav": true,
+		"mp3": true,
+	}
+
+	if _, ok := format[request.ResponseFormat]; ok {
+		ttsRequest.Payload.Parameters.Format = request.ResponseFormat
+	}
+
+	if 0.5 <= request.Speed && request.Speed <= 2 {
+		ttsRequest.Payload.Parameters.Rate = request.Speed
+	}
+
+	return &ttsRequest
+}
+
 func EmbeddingHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
 	var aliResponse EmbeddingResponse
 	err := json.NewDecoder(resp.Body).Decode(&aliResponse)

relay/adaptor/anthropic/adaptor.go

@@ -48,6 +48,13 @@ func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error)
 	return request, nil
 }
 
+func (a *Adaptor) ConvertTextToSpeechRequest(request *model.TextToSpeechRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
 	return adaptor.DoRequestHelper(a, c, meta, requestBody)
 }

relay/adaptor/aws/adapter.go

@@ -57,6 +57,13 @@ func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error)
 	return request, nil
 }
 
+func (a *Adaptor) ConvertTextToSpeechRequest(request *model.TextToSpeechRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
 	return nil, nil
 }

relay/adaptor/baidu/adaptor.go

@@ -116,6 +116,13 @@ func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error)
 	return request, nil
 }
 
+func (a *Adaptor) ConvertTextToSpeechRequest(request *model.TextToSpeechRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
 	return adaptor.DoRequestHelper(a, c, meta, requestBody)
 }

relay/adaptor/cloudflare/adaptor.go

@@ -44,6 +44,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	return ConvertRequest(*request), nil
 }
 
+func (a *Adaptor) ConvertTextToSpeechRequest(request *model.TextToSpeechRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
 	return adaptor.DoRequestHelper(a, c, meta, requestBody)
 }

relay/adaptor/cohere/adaptor.go

@@ -42,6 +42,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	return ConvertRequest(*request), nil
 }
 
+func (a *Adaptor) ConvertTextToSpeechRequest(request *model.TextToSpeechRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
 	return adaptor.DoRequestHelper(a, c, meta, requestBody)
 }

relay/adaptor/coze/adaptor.go

@@ -45,6 +45,13 @@ func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error)
 	return request, nil
 }
 
+func (a *Adaptor) ConvertTextToSpeechRequest(request *model.TextToSpeechRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
 	return adaptor.DoRequestHelper(a, c, meta, requestBody)
 }

relay/adaptor/deepl/adaptor.go

@@ -46,6 +46,13 @@ func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error)
 	return request, nil
 }
 
+func (a *Adaptor) ConvertTextToSpeechRequest(request *model.TextToSpeechRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
 	return adaptor.DoRequestHelper(a, c, meta, requestBody)
 }

relay/adaptor/gemini/adaptor.go

@@ -66,6 +66,13 @@ func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error)
 	return request, nil
 }
 
+func (a *Adaptor) ConvertTextToSpeechRequest(request *model.TextToSpeechRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
 	return channelhelper.DoRequestHelper(a, c, meta, requestBody)
 }

relay/adaptor/interface.go

@@ -14,6 +14,7 @@ type Adaptor interface {
 	SetupRequestHeader(c *gin.Context, req *http.Request, meta *meta.Meta) error
 	ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error)
 	ConvertImageRequest(request *model.ImageRequest) (any, error)
+	ConvertTextToSpeechRequest(request *model.TextToSpeechRequest) (any, error)
 	DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error)
 	DoResponse(c *gin.Context, resp *http.Response, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode)
 	GetModelList() []string

relay/adaptor/ollama/adaptor.go

@@ -55,6 +55,13 @@ func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error)
 	return request, nil
 }
 
+func (a *Adaptor) ConvertTextToSpeechRequest(request *model.TextToSpeechRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
 	return adaptor.DoRequestHelper(a, c, meta, requestBody)
 }