Merge pull request #52 from seymourtang/convoai-1.6

feat:Add support for convo AI 1.6

Author: seymourtang

examples/convoai/README.md

@@ -37,6 +37,25 @@ export CONVOAI_TTS_ELEVENLABS_MODEL_ID=<Your tts elevenLabs model id>
 export CONVOAI_TTS_ELEVENLABS_VOICE_ID=<Your tts elevenLabs voice id>
 ```
 
+### cartesia
+
+```bash
+export CONVOAI_TTS_CARTESIA_API_KEY=<Your tts cartesia api key>
+export CONVOAI_TTS_CARTESIA_MODEL_ID=<Your tts cartesia model id>
+export CONVOAI_TTS_CARTESIA_VOICE_MODE=<Your tts cartesia voice mode>
+export CONVOAI_TTS_CARTESIA_VOICE_ID=<Your tts cartesia voice id>
+```
+
+### openai
+
+```bash
+export CONVOAI_TTS_OPENAI_API_KEY=<Your tts openai api key>
+export CONVOAI_TTS_OPENAI_MODEL=<Your tts openai model>
+export CONVOAI_TTS_OPENAI_VOICE=<Your tts openai voice>
+export CONVOAI_TTS_OPENAI_INSTRUCTIONS=<Your tts openai instructions>
+export CONVOAI_TTS_OPENAI_SPEED=<Your tts openai speed>
+```
+
 ## Execution
 
 Run the sample project with the following command:
@@ -49,5 +68,7 @@ go run main.go --ttsVendor=<ttsVendor> --serviceRegion=2
 
 -   `microsoft`
 -   `elevenLabs`
+-   `cartesia`
+-   `openai`
 
 Choose the appropriate TTS provider based on your requirements.

examples/convoai/main.go

@@ -35,7 +35,7 @@ func main() {
 		panic("BASIC_AUTH_PASSWORD is required")
 	}
 
-	ttsVendor := flag.String("ttsVendor", "", "tts vendor, e.g. bytedance,microsoft,tencent,minimax,elevenLabs")
+	ttsVendor := flag.String("ttsVendor", "", "tts vendor, e.g. bytedance,microsoft,tencent,minimax,elevenLabs,cartesia,openai")
 	serviceRegion := flag.Int("serviceRegion", 0, "service region, e.g. 1: ChineseMainland, 2: Global")
 	flag.Parse()
 
@@ -54,10 +54,6 @@ func main() {
 
 		s.RunWithBytedanceTTS()
 	case "microsoft":
-		if *serviceRegion == 2 {
-			log.Fatalln("Microsoft TTS is not supported in ChineseMainland service region")
-		}
-
 		s.RunWithMicrosoftTTS()
 	case "tencent":
 		if *serviceRegion == 2 {
@@ -72,11 +68,23 @@ func main() {
 
 		s.RunWithMinimaxTTS()
 	case "elevenLabs":
-		if *serviceRegion == 2 {
+		if *serviceRegion == 1 {
 			log.Fatalln("ElevenLabs TTS is not supported in ChineseMainland service region")
 		}
 
 		s.RunWithElevenLabsTTS()
+	case "cartesia":
+		if *serviceRegion == 1 {
+			log.Fatalln("Cartesia TTS is not supported in ChineseMainland service region")
+		}
+
+		s.RunWithCartesiaTTS()
+	case "openai":
+		if *serviceRegion == 1 {
+			log.Fatalln("OpenAI TTS is not supported in ChineseMainland service region")
+		}
+
+		s.RunWithOpenAITTS()
 	default:
 		log.Fatalln("Invalid tts vendor")
 	}

examples/convoai/service/service.go

@@ -4,6 +4,7 @@ import (
 	"context"
 	"log"
 	"os"
+	"strconv"
 	"time"
 
 	"github.com/AgoraIO-Community/agora-rest-client-go/agora/auth"
@@ -34,7 +35,7 @@ func (s *Service) SetCredential(username string, password string) {
 	s.credential = auth.NewBasicAuthCredential(username, password)
 }
 
-func (s *Service) RunWithCustomTTS(ttsVendor req.TTSVendor, ttsParam req.TTSVendorParamsInterface) {
+func (s *Service) RunWithCustomTTS(ttsParam req.TTSVendorParamsInterface) {
 	ctx := context.Background()
 	config := &convoai.Config{
 		AppID:         s.appId,
@@ -114,7 +115,7 @@ func (s *Service) RunWithCustomTTS(ttsVendor req.TTSVendor, ttsParam req.TTSVend
 			GreetingMessage: "Hello,how can I help you?",
 		},
 		TTS: &req.JoinPropertiesTTSBody{
-			Vendor: ttsVendor,
+			Vendor: ttsParam.GetVendorType(),
 			Params: ttsParam,
 		},
 		Vad: &req.JoinPropertiesVadBody{
@@ -320,7 +321,7 @@ func (s *Service) RunWithBytedanceTTS() {
 		PitchRatio:  1.0,
 		Emotion:     "happy",
 	}
-	s.RunWithCustomTTS(req.BytedanceTTSVendor, ttsParam)
+	s.RunWithCustomTTS(ttsParam)
 }
 
 func (s *Service) RunWithTencentTTS() {
@@ -349,7 +350,7 @@ func (s *Service) RunWithTencentTTS() {
 		EmotionCategory:  "happy",
 		EmotionIntensity: 100,
 	}
-	s.RunWithCustomTTS(req.TencentTTSVendor, ttsParam)
+	s.RunWithCustomTTS(ttsParam)
 }
 
 func (s *Service) RunWithMinimaxTTS() {
@@ -384,7 +385,7 @@ func (s *Service) RunWithMinimaxTTS() {
 		},
 	}
 
-	s.RunWithCustomTTS(req.MinimaxTTSVendor, ttsParam)
+	s.RunWithCustomTTS(ttsParam)
 }
 
 func (s *Service) RunWithMicrosoftTTS() {
@@ -412,7 +413,7 @@ func (s *Service) RunWithMicrosoftTTS() {
 		SampleRate: 24000,
 	}
 
-	s.RunWithCustomTTS(req.MicrosoftTTSVendor, ttsParam)
+	s.RunWithCustomTTS(ttsParam)
 }
 
 func (s *Service) RunWithElevenLabsTTS() {
@@ -437,5 +438,80 @@ func (s *Service) RunWithElevenLabsTTS() {
 		VoiceId: ttsVoiceId,
 	}
 
-	s.RunWithCustomTTS(req.ElevenLabsTTSVendor, ttsParam)
+	s.RunWithCustomTTS(ttsParam)
+}
+
+func (s *Service) RunWithCartesiaTTS() {
+	ttsApiKey := os.Getenv("CONVOAI_TTS_CARTESIA_API_KEY")
+	if ttsApiKey == "" {
+		log.Fatalln("CONVOAI_TTS_CARTESIA_API_KEY is required")
+	}
+
+	ttsModelId := os.Getenv("CONVOAI_TTS_CARTESIA_MODEL_ID")
+	if ttsModelId == "" {
+		log.Fatalln("CONVOAI_TTS_CARTESIA_MODEL_ID is required")
+	}
+
+	ttsVoiceMode := os.Getenv("CONVOAI_TTS_CARTESIA_VOICE_MODE")
+	if ttsVoiceMode == "" {
+		log.Fatalln("CONVOAI_TTS_CARTESIA_VOICE_MODE is required")
+	}
+
+	ttsVoiceId := os.Getenv("CONVOAI_TTS_CARTESIA_VOICE_ID")
+	if ttsVoiceId == "" {
+		log.Fatalln("CONVOAI_TTS_CARTESIA_VOICE_ID is required")
+	}
+
+	ttsParam := req.TTSCartesiaVendorParams{
+		APIKey:  ttsApiKey,
+		ModelId: ttsModelId,
+		Voice: &req.TTSCartesiaVendorVoice{
+			Mode: ttsVoiceMode,
+			Id:   ttsVoiceId,
+		},
+	}
+
+	s.RunWithCustomTTS(ttsParam)
+}
+
+func (s *Service) RunWithOpenAITTS() {
+	ttsApiKey := os.Getenv("CONVOAI_TTS_OPENAI_API_KEY")
+	if ttsApiKey == "" {
+		log.Fatalln("CONVOAI_TTS_OPENAI_API_KEY is required")
+	}
+
+	ttsModel := os.Getenv("CONVOAI_TTS_OPENAI_MODEL")
+	if ttsModel == "" {
+		log.Fatalln("CONVOAI_TTS_OPENAI_MODEL is required")
+	}
+
+	ttsVoice := os.Getenv("CONVOAI_TTS_OPENAI_VOICE")
+	if ttsVoice == "" {
+		log.Fatalln("CONVOAI_TTS_OPENAI_VOICE is required")
+	}
+
+	ttsInstructions := os.Getenv("CONVOAI_TTS_OPENAI_INSTRUCTIONS")
+	if ttsInstructions == "" {
+		log.Fatalln("CONVOAI_TTS_OPENAI_INSTRUCTIONS is required")
+	}
+
+	ttsSpeed := os.Getenv("CONVOAI_TTS_OPENAI_SPEED")
+	if ttsSpeed == "" {
+		log.Fatalln("CONVOAI_TTS_OPENAI_SPEED is required")
+	}
+
+	speed, err := strconv.ParseFloat(ttsSpeed, 32)
+	if err != nil {
+		log.Fatalln("CONVOAI_TTS_OPENAI_SPEED is not a valid float")
+	}
+
+	ttsParam := req.TTSOpenAIVendorParams{
+		APIKey:       ttsApiKey,
+		Model:        ttsModel,
+		Voice:        ttsVoice,
+		Instructions: ttsInstructions,
+		Speed:        float32(speed),
+	}
+
+	s.RunWithCustomTTS(ttsParam)
 }