XEngineers/CamTalk
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request 'fix: 修复 deploy 工作流缺少 Node.js 导致 checkout 失败的问题' #72

Merged
huanghaosheng merged 61 commits from develop into main 2026-06-14 14:37:27 +08:00
Conversation 0 Commits 61 Files Changed 52 +269 -151
18 changed files with 269 additions and 151 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Showing only changes of commit 563ca12790 - Show all commits

35
backend/cmd/server/main.go
View File

@@ -22,6 +22,10 @@ import (
"github.com/hhs/camtalk/internal/ws"
)
// Version 通过构建时 -ldflags 注入，如：
// go build -ldflags "-X main.Version=v1.0.0" ./cmd/server
var Version string
var startTime = time.Now()
func main() {
@@ -43,7 +47,10 @@ func main() {
// 初始化 Session Manager（MVP 默认内存实现）
var sessionMgr session.Manager
// TODO: 当 Redis 配置非空时切换为 RedisManager
sessionMgr = session.NewMemoryManager(30*time.Minute, 20)
sessionMgr = session.NewMemoryManager(
time.Duration(cfg.Session.TTL)*time.Minute,
cfg.Session.MaxHistory,
)
defer sessionMgr.(*session.MemoryManager).Stop()
// 初始化 AI 服务
@@ -60,27 +67,27 @@ func main() {
var sttService stt.Service
switch strings.ToLower(cfg.AI.STT.Provider) {
case "mimo", "xiaomi":
sttService = stt.NewMiMoService(cfg.AI.STT.APIKey, cfg.AI.STT.Model, cfg.AI.STT.Endpoint, logger.Log)
sttService = stt.NewMiMoService(cfg.AI.STT.APIKey, cfg.AI.STT.Model, cfg.AI.STT.Endpoint, cfg.AI.STT.Timeout, logger.Log)
logger.Log.Infow("STT service initialized", "provider", "mimo", "model", cfg.AI.STT.Model, "endpoint", cfg.AI.STT.Endpoint)
default:
sttService = stt.NewDeepgramService(cfg.AI.STT.APIKey, cfg.AI.STT.Model, cfg.AI.STT.Endpoint, logger.Log)
sttService = stt.NewDeepgramService(cfg.AI.STT.APIKey, cfg.AI.STT.Model, cfg.AI.STT.Endpoint, cfg.AI.STT.Timeout, logger.Log)
logger.Log.Infow("STT service initialized", "provider", "deepgram", "model", cfg.AI.STT.Model)
}
llmService := llm.NewOpenAIService(cfg.AI.LLM.APIKey, cfg.AI.LLM.Model, cfg.AI.LLM.Endpoint, cfg.AI.LLM.Timeout, logger.Log)
llmService := llm.NewOpenAIService(cfg.AI.LLM.APIKey, cfg.AI.LLM.Model, cfg.AI.LLM.Endpoint, cfg.AI.LLM.Timeout, cfg.AI.LLM.HTTPClientTimeout, logger.Log)
logger.Log.Infow("LLM service initialized", "provider", cfg.AI.LLM.Provider, "model", cfg.AI.LLM.Model, "endpoint", cfg.AI.LLM.Endpoint, "timeout", cfg.AI.LLM.Timeout)
var ttsService tts.Service
switch strings.ToLower(cfg.AI.TTS.Provider) {
case "mimo", "xiaomi":
ttsService = tts.NewMiMoService(cfg.AI.TTS.APIKey, cfg.AI.TTS.Model, cfg.AI.TTS.Voice, cfg.AI.TTS.Endpoint, cfg.AI.TTS.Timeout, logger.Log)
ttsService = tts.NewMiMoService(cfg.AI.TTS.APIKey, cfg.AI.TTS.Model, cfg.AI.TTS.Voice, cfg.AI.TTS.Endpoint, cfg.AI.TTS.Timeout, cfg.AI.TTS.HTTPClientTimeout, logger.Log)
logger.Log.Infow("TTS service initialized", "provider", "mimo", "model", cfg.AI.TTS.Model, "voice", cfg.AI.TTS.Voice, "endpoint", cfg.AI.TTS.Endpoint)
default:
ttsService = tts.NewOpenAIService(cfg.AI.TTS.APIKey, cfg.AI.TTS.Model, cfg.AI.TTS.Voice, cfg.AI.TTS.Endpoint, cfg.AI.TTS.Speed, cfg.AI.TTS.Timeout, logger.Log)
ttsService = tts.NewOpenAIService(cfg.AI.TTS.APIKey, cfg.AI.TTS.Model, cfg.AI.TTS.Voice, cfg.AI.TTS.Endpoint, cfg.AI.TTS.Speed, cfg.AI.TTS.Timeout, cfg.AI.TTS.HTTPClientTimeout, logger.Log)
logger.Log.Infow("TTS service initialized", "provider", "openai", "model", cfg.AI.TTS.Model, "voice", cfg.AI.TTS.Voice, "speed", cfg.AI.TTS.Speed)
}
// 初始化 Orchestrator
orch := orchestrator.New(sttService, llmService, ttsService, sessionMgr, cfg.AI.LLM.Model, cfg.AI.TTS.Voice, cfg.AI.TTS.Speed)
orch := orchestrator.New(sttService, llmService, ttsService, sessionMgr, cfg)
// Gin 模式
if cfg.App.Env == "prod" {
@@ -93,7 +100,7 @@ func main() {
// REST API
apiGroup := r.Group("/api")
{
apiGroup.GET("/health", healthHandler(sessionMgr))
apiGroup.GET("/health", healthHandler(sessionMgr, cfg))
}
// Session REST 端点
@@ -101,7 +108,7 @@ func main() {
sessionHandler.RegisterRoutes(apiGroup)
// WebSocket
r.GET("/ws", ws.ServeWS(sessionMgr, orch))
r.GET("/ws", ws.ServeWS(sessionMgr, orch, cfg))
// HTTP Server
srv := &http.Server{
@@ -125,7 +132,7 @@ func main() {
<-ctx.Done()
logger.Log.Info("shutting down...")
shutdownCtx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
shutdownCtx, cancel := context.WithTimeout(context.Background(), time.Duration(cfg.Server.ShutdownTimeout)*time.Second)
defer cancel()
if err := srv.Shutdown(shutdownCtx); err != nil {
@@ -135,11 +142,15 @@ func main() {
}
// healthHandler 健康检查。
func healthHandler(sessionMgr session.Manager) gin.HandlerFunc {
func healthHandler(sessionMgr session.Manager, cfg *config.Config) gin.HandlerFunc {
return func(c *gin.Context) {
version := Version
if version == "" {
version = cfg.App.Version
}
c.JSON(200, gin.H{
"status": "ok",
"version": "0.1.0",
"version": version,
"uptime_seconds": int(time.Since(startTime).Seconds()),
"active_sessions": sessionMgr.ActiveCount(),
})

15
backend/internal/ai/llm/openai.go
View File

@@ -26,24 +26,23 @@ type OpenAIService struct {
}
// NewOpenAIService 创建 OpenAI LLM 服务。
func NewOpenAIService(apiKey, model, endpoint string, timeoutSec int, logger *zap.SugaredLogger) *OpenAIService {
if model == "" {
model = "gpt-4o"
}
if endpoint == "" {
endpoint = "https://api.openai.com/v1"
}
// model、endpoint 由 config 层保证非空。
func NewOpenAIService(apiKey, model, endpoint string, timeoutSec, httpClientTimeoutSec int, logger *zap.SugaredLogger) *OpenAIService {
timeout := time.Duration(timeoutSec) * time.Second
if timeout <= 0 {
timeout = 10 * time.Second
}
httpClientTimeout := time.Duration(httpClientTimeoutSec) * time.Second
if httpClientTimeout <= 0 {
httpClientTimeout = 60 * time.Second
}
return &OpenAIService{
apiKey: apiKey,
model: model,
endpoint: endpoint,
timeout: timeout,
logger: logger,
client: &http.Client{Timeout: 60 * time.Second}, // HTTP client timeout > LLM timeout
client: &http.Client{Timeout: httpClientTimeout},
}
}

12
backend/internal/ai/llm/openai_test.go
View File

@@ -53,7 +53,7 @@ func TestOpenAIService_ChatStream_Success(t *testing.T) {
})
defer srv.Close()
svc := NewOpenAIService("test-key", "gpt-4o", srv.URL, 10, zap.NewNop().Sugar())
svc := NewOpenAIService("test-key", "gpt-4o", srv.URL, 10, 60, zap.NewNop().Sugar())
ch, err := svc.ChatStream(context.Background(), Request{
Text: "这是什么？",
@@ -99,7 +99,7 @@ func TestOpenAIService_ChatStream_WithImage(t *testing.T) {
})
defer srv.Close()
svc := NewOpenAIService("test-key", "gpt-4o", srv.URL, 10, zap.NewNop().Sugar())
svc := NewOpenAIService("test-key", "gpt-4o", srv.URL, 10, 60, zap.NewNop().Sugar())
ch, err := svc.ChatStream(context.Background(), Request{
Image: []byte("fake-jpeg-data"),
@@ -123,7 +123,7 @@ func TestOpenAIService_ChatStream_WithHistory(t *testing.T) {
})
defer srv.Close()
svc := NewOpenAIService("test-key", "gpt-4o", srv.URL, 10, zap.NewNop().Sugar())
svc := NewOpenAIService("test-key", "gpt-4o", srv.URL, 10, 60, zap.NewNop().Sugar())
ch, err := svc.ChatStream(context.Background(), Request{
Text: "继续",
@@ -149,7 +149,7 @@ func TestOpenAIService_ChatStream_APIError(t *testing.T) {
})
defer srv.Close()
svc := NewOpenAIService("bad-key", "gpt-4o", srv.URL, 10, zap.NewNop().Sugar())
svc := NewOpenAIService("bad-key", "gpt-4o", srv.URL, 10, 60, zap.NewNop().Sugar())
_, err := svc.ChatStream(context.Background(), Request{
Text: "test",
@@ -172,7 +172,7 @@ func TestOpenAIService_ChatStream_Timeout(t *testing.T) {
})
defer srv.Close()
svc := NewOpenAIService("test-key", "gpt-4o", srv.URL, 1, zap.NewNop().Sugar()) // 1s timeout
svc := NewOpenAIService("test-key", "gpt-4o", srv.URL, 1, 60, zap.NewNop().Sugar()) // 1s timeout
ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
defer cancel()
@@ -204,7 +204,7 @@ func TestOpenAIService_ChatStream_UsageInResponse(t *testing.T) {
})
defer srv.Close()
svc := NewOpenAIService("test-key", "gpt-4o", srv.URL, 10, zap.NewNop().Sugar())
svc := NewOpenAIService("test-key", "gpt-4o", srv.URL, 10, 60, zap.NewNop().Sugar())
ch, err := svc.ChatStream(context.Background(), Request{Text: "test"})
if err != nil {

17
backend/internal/ai/stt/deepgram.go
View File

@@ -18,21 +18,22 @@ type DeepgramService struct {
apiKey string
model string
endpoint string
timeout time.Duration
logger *zap.SugaredLogger
}
// NewDeepgramService 创建 Deepgram STT 服务。
func NewDeepgramService(apiKey, model, endpoint string, logger *zap.SugaredLogger) *DeepgramService {
if model == "" {
model = "nova-2"
}
if endpoint == "" {
endpoint = "wss://api.deepgram.com/v1/listen"
// model、endpoint 由 config 层保证非空，timeoutSec 为 0 时默认 5 秒。
func NewDeepgramService(apiKey, model, endpoint string, timeoutSec int, logger *zap.SugaredLogger) *DeepgramService {
timeout := time.Duration(timeoutSec) * time.Second
if timeout <= 0 {
timeout = 5 * time.Second
}
return &DeepgramService{
apiKey: apiKey,
model: model,
endpoint: endpoint,
timeout: timeout,
logger: logger,
}
}
@@ -57,8 +58,8 @@ func (d *DeepgramService) Recognize(ctx context.Context, audio []byte, opts Opti
// 构建 WebSocket URL，附带查询参数
wsURL := d.buildURL(opts)
// 5 秒总超时
ctx, cancel := context.WithTimeout(ctx, 5*time.Second)
// 总超时
ctx, cancel := context.WithTimeout(ctx, d.timeout)
defer cancel()
// 建立 WebSocket 连接

12
backend/internal/ai/stt/deepgram_test.go
View File

@@ -74,7 +74,7 @@ func TestDeepgramService_Recognize_Success(t *testing.T) {
})
defer srv.Close()
svc := NewDeepgramService("test-key", "", wsToWss(srv.URL)+"/v1/listen", zap.NewNop().Sugar())
svc := NewDeepgramService("test-key", "", wsToWss(srv.URL)+"/v1/listen", 0, zap.NewNop().Sugar())
text, err := svc.Recognize(context.Background(), []byte("fake-pcm-audio"), Options{
Encoding: "pcm_s16le",
@@ -90,7 +90,7 @@ func TestDeepgramService_Recognize_Success(t *testing.T) {
}
func TestDeepgramService_Recognize_EmptyAudio(t *testing.T) {
svc := NewDeepgramService("test-key", "", "ws://localhost", zap.NewNop().Sugar())
svc := NewDeepgramService("test-key", "", "ws://localhost", 0, zap.NewNop().Sugar())
_, err := svc.Recognize(context.Background(), nil, Options{})
if err == nil {
t.Fatal("Recognize() with empty audio should return error")
@@ -98,7 +98,7 @@ func TestDeepgramService_Recognize_EmptyAudio(t *testing.T) {
}
func TestDeepgramService_Recognize_ConnectError(t *testing.T) {
svc := NewDeepgramService("test-key", "", "ws://localhost:1", zap.NewNop().Sugar())
svc := NewDeepgramService("test-key", "", "ws://localhost:1", 0, zap.NewNop().Sugar())
ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
defer cancel()
@@ -117,7 +117,7 @@ func TestDeepgramService_Recognize_Timeout(t *testing.T) {
})
defer srv.Close()
svc := NewDeepgramService("test-key", "", wsToWss(srv.URL)+"/v1/listen", zap.NewNop().Sugar())
svc := NewDeepgramService("test-key", "", wsToWss(srv.URL)+"/v1/listen", 0, zap.NewNop().Sugar())
ctx, cancel := context.WithTimeout(context.Background(), 1*time.Second)
defer cancel()
@@ -147,7 +147,7 @@ func TestDeepgramService_Recognize_MultipleFinals(t *testing.T) {
})
defer srv.Close()
svc := NewDeepgramService("test-key", "", wsToWss(srv.URL)+"/v1/listen", zap.NewNop().Sugar())
svc := NewDeepgramService("test-key", "", wsToWss(srv.URL)+"/v1/listen", 0, zap.NewNop().Sugar())
text, err := svc.Recognize(context.Background(), []byte("audio"), Options{})
if err != nil {
@@ -159,7 +159,7 @@ func TestDeepgramService_Recognize_MultipleFinals(t *testing.T) {
}
func TestDeepgramService_buildURL(t *testing.T) {
svc := NewDeepgramService("key", "", "wss://api.deepgram.com/v1/listen", zap.NewNop().Sugar())
svc := NewDeepgramService("key", "", "wss://api.deepgram.com/v1/listen", 0, zap.NewNop().Sugar())
tests := []struct {
name string

15
backend/internal/ai/stt/mimo.go
View File

@@ -21,21 +21,22 @@ type MiMoService struct {
apiKey string
model string
endpoint string
timeout time.Duration
logger *zap.SugaredLogger
}
// NewMiMoService 创建 MiMo STT 服务。
func NewMiMoService(apiKey, model, endpoint string, logger *zap.SugaredLogger) *MiMoService {
if model == "" {
model = "mimo-v2.5-asr"
}
if endpoint == "" {
endpoint = "https://api.xiaomimimo.com/v1"
// model、endpoint 由 config 层保证非空，timeoutSec 为 0 时默认 10 秒。
func NewMiMoService(apiKey, model, endpoint string, timeoutSec int, logger *zap.SugaredLogger) *MiMoService {
timeout := time.Duration(timeoutSec) * time.Second
if timeout <= 0 {
timeout = 10 * time.Second
}
return &MiMoService{
apiKey: apiKey,
model: model,
endpoint: endpoint,
timeout: timeout,
logger: logger,
}
}
@@ -126,7 +127,7 @@ func (m *MiMoService) Recognize(ctx context.Context, audio []byte, opts Options)
url := strings.TrimRight(m.endpoint, "/") + "/chat/completions"
ctx, cancel := context.WithTimeout(ctx, 10*time.Second)
ctx, cancel := context.WithTimeout(ctx, m.timeout)
defer cancel()
req, err := http.NewRequestWithContext(ctx, http.MethodPost, url, bytes.NewReader(body))

2
backend/internal/ai/stt/mimo_test.go
View File

@@ -14,7 +14,7 @@ import (
func newTestMiMoService(handler http.HandlerFunc) (*MiMoService, *httptest.Server) {
srv := httptest.NewServer(handler)
s := NewMiMoService("test-key", "mimo-v2.5-asr", srv.URL, zap.NewNop().Sugar())
s := NewMiMoService("test-key", "mimo-v2.5-asr", srv.URL, 0, zap.NewNop().Sugar())
return s, srv
}

18
backend/internal/ai/tts/mimo.go
View File

@@ -27,20 +27,16 @@ type MiMoService struct {
}
// NewMiMoService 创建 MiMo TTS 服务。
func NewMiMoService(apiKey, model, voice, endpoint string, timeoutSec int, logger *zap.SugaredLogger) *MiMoService {
if model == "" {
model = "mimo-v2.5-tts"
}
if voice == "" {
voice = "冰糖"
}
if endpoint == "" {
endpoint = "https://api.xiaomimimo.com/v1"
}
// model、voice、endpoint 由 config 层保证非空。
func NewMiMoService(apiKey, model, voice, endpoint string, timeoutSec, httpClientTimeoutSec int, logger *zap.SugaredLogger) *MiMoService {
timeout := time.Duration(timeoutSec) * time.Second
if timeout <= 0 {
timeout = 5 * time.Second
}
httpClientTimeout := time.Duration(httpClientTimeoutSec) * time.Second
if httpClientTimeout <= 0 {
httpClientTimeout = 30 * time.Second
}
return &MiMoService{
apiKey: apiKey,
model: model,
@@ -48,7 +44,7 @@ func NewMiMoService(apiKey, model, voice, endpoint string, timeoutSec int, logge
endpoint: endpoint,
timeout: timeout,
logger: logger,
client: &http.Client{Timeout: 30 * time.Second},
client: &http.Client{Timeout: httpClientTimeout},
}
}

18
backend/internal/ai/tts/mimo_test.go
View File

@@ -93,7 +93,7 @@ func TestMiMoService_SynthesizeStream_Success(t *testing.T) {
})
defer srv.Close()
svc := NewMiMoService("test-key", "", "冰糖", srv.URL, 5, zap.NewNop().Sugar())
svc := NewMiMoService("test-key", "mimo-v2.5-tts", "冰糖", srv.URL, 5, 30, zap.NewNop().Sugar())
textStream := sendSentences("你好", "世界", "！")
@@ -145,7 +145,7 @@ func TestMiMoService_SynthesizeStream_APIError(t *testing.T) {
})
defer srv.Close()
svc := NewMiMoService("test-key", "", "冰糖", srv.URL, 5, zap.NewNop().Sugar())
svc := NewMiMoService("test-key", "mimo-v2.5-tts", "冰糖", srv.URL, 5, 30, zap.NewNop().Sugar())
textStream := sendSentences("你好")
@@ -178,7 +178,7 @@ func TestMiMoService_SynthesizeStream_Timeout(t *testing.T) {
defer srv.Close()
// 1 秒超时
svc := NewMiMoService("test-key", "", "冰糖", srv.URL, 1, zap.NewNop().Sugar())
svc := NewMiMoService("test-key", "mimo-v2.5-tts", "冰糖", srv.URL, 1, 30, zap.NewNop().Sugar())
textStream := sendSentences("很长的句子")
@@ -214,7 +214,7 @@ func TestMiMoService_SynthesizeStream_EmptyText(t *testing.T) {
})
defer srv.Close()
svc := NewMiMoService("test-key", "", "冰糖", srv.URL, 5, zap.NewNop().Sugar())
svc := NewMiMoService("test-key", "mimo-v2.5-tts", "冰糖", srv.URL, 5, 30, zap.NewNop().Sugar())
// 空句子应该被跳过
textStream := sendSentences("", "你好", "")
@@ -248,7 +248,7 @@ func TestMiMoService_SynthesizeStream_ContextCancelled(t *testing.T) {
})
defer srv.Close()
svc := NewMiMoService("test-key", "", "冰糖", srv.URL, 5, zap.NewNop().Sugar())
svc := NewMiMoService("test-key", "mimo-v2.5-tts", "冰糖", srv.URL, 5, 30, zap.NewNop().Sugar())
textStream := make(chan string, 3)
textStream <- "第一句"
@@ -290,7 +290,7 @@ func TestMiMoService_SynthesizeStream_PartialFailure(t *testing.T) {
})
defer srv.Close()
svc := NewMiMoService("test-key", "", "冰糖", srv.URL, 5, zap.NewNop().Sugar())
svc := NewMiMoService("test-key", "mimo-v2.5-tts", "冰糖", srv.URL, 5, 30, zap.NewNop().Sugar())
textStream := sendSentences("第一句", "第二句", "第三句")
@@ -329,7 +329,7 @@ func TestMiMoService_SynthesizeStream_CustomVoice(t *testing.T) {
})
defer srv.Close()
svc := NewMiMoService("test-key", "", "冰糖", srv.URL, 5, zap.NewNop().Sugar())
svc := NewMiMoService("test-key", "mimo-v2.5-tts", "冰糖", srv.URL, 5, 30, zap.NewNop().Sugar())
textStream := sendSentences("你好")
@@ -360,7 +360,7 @@ func TestMiMoService_SynthesizeStream_DefaultVoice(t *testing.T) {
defer srv.Close()
// 不指定 voice
svc := NewMiMoService("test-key", "", "", srv.URL, 5, zap.NewNop().Sugar())
svc := NewMiMoService("test-key", "mimo-v2.5-tts", "冰糖", srv.URL, 5, 30, zap.NewNop().Sugar())
textStream := sendSentences("你好")
@@ -381,7 +381,7 @@ func TestMiMoService_SynthesizeStream_EmptyAudioData(t *testing.T) {
})
defer srv.Close()
svc := NewMiMoService("test-key", "", "冰糖", srv.URL, 5, zap.NewNop().Sugar())
svc := NewMiMoService("test-key", "mimo-v2.5-tts", "冰糖", srv.URL, 5, 30, zap.NewNop().Sugar())
textStream := sendSentences("你好")

18
backend/internal/ai/tts/openai.go
View File

@@ -25,23 +25,19 @@ type OpenAIService struct {
}
// NewOpenAIService 创建 OpenAI TTS 服务。
func NewOpenAIService(apiKey, model, voice, endpoint string, speed float64, timeoutSec int, logger *zap.SugaredLogger) *OpenAIService {
if model == "" {
model = "tts-1"
}
if voice == "" {
voice = "alloy"
}
// model、voice、endpoint 由 config 层保证非空。
func NewOpenAIService(apiKey, model, voice, endpoint string, speed float64, timeoutSec, httpClientTimeoutSec int, logger *zap.SugaredLogger) *OpenAIService {
if speed <= 0 {
speed = 1.0
}
if endpoint == "" {
endpoint = "https://api.openai.com/v1"
}
timeout := time.Duration(timeoutSec) * time.Second
if timeout <= 0 {
timeout = 5 * time.Second
}
httpClientTimeout := time.Duration(httpClientTimeoutSec) * time.Second
if httpClientTimeout <= 0 {
httpClientTimeout = 30 * time.Second
}
return &OpenAIService{
apiKey: apiKey,
model: model,
@@ -50,7 +46,7 @@ func NewOpenAIService(apiKey, model, voice, endpoint string, speed float64, time
endpoint: endpoint,
timeout: timeout,
logger: logger,
client: &http.Client{Timeout: 30 * time.Second},
client: &http.Client{Timeout: httpClientTimeout},
}
}

14
backend/internal/ai/tts/openai_test.go
View File

@@ -58,7 +58,7 @@ func TestOpenAIService_SynthesizeStream_Success(t *testing.T) {
})
defer srv.Close()
svc := NewOpenAIService("test-key", "", "alloy", srv.URL, 1.0, 5, zap.NewNop().Sugar())
svc := NewOpenAIService("test-key", "tts-1", "alloy", srv.URL, 1.0, 5, 30, zap.NewNop().Sugar())
textStream := sendSentences("你好", "世界", "！")
@@ -110,7 +110,7 @@ func TestOpenAIService_SynthesizeStream_APIError(t *testing.T) {
})
defer srv.Close()
svc := NewOpenAIService("test-key", "", "alloy", srv.URL, 1.0, 5, zap.NewNop().Sugar())
svc := NewOpenAIService("test-key", "tts-1", "alloy", srv.URL, 1.0, 5, 30, zap.NewNop().Sugar())
textStream := sendSentences("你好")
@@ -142,7 +142,7 @@ func TestOpenAIService_SynthesizeStream_Timeout(t *testing.T) {
defer srv.Close()
// 1 秒超时
svc := NewOpenAIService("test-key", "", "alloy", srv.URL, 1.0, 1, zap.NewNop().Sugar())
svc := NewOpenAIService("test-key", "tts-1", "alloy", srv.URL, 1.0, 1, 30, zap.NewNop().Sugar())
textStream := sendSentences("很长的句子")
@@ -177,7 +177,7 @@ func TestOpenAIService_SynthesizeStream_EmptyText(t *testing.T) {
})
defer srv.Close()
svc := NewOpenAIService("test-key", "", "alloy", srv.URL, 1.0, 5, zap.NewNop().Sugar())
svc := NewOpenAIService("test-key", "tts-1", "alloy", srv.URL, 1.0, 5, 30, zap.NewNop().Sugar())
// 空句子应该被跳过
textStream := sendSentences("", "你好", "")
@@ -210,7 +210,7 @@ func TestOpenAIService_SynthesizeStream_ContextCancelled(t *testing.T) {
})
defer srv.Close()
svc := NewOpenAIService("test-key", "", "alloy", srv.URL, 1.0, 5, zap.NewNop().Sugar())
svc := NewOpenAIService("test-key", "tts-1", "alloy", srv.URL, 1.0, 5, 30, zap.NewNop().Sugar())
// 发送多个句子，但在第一个后取消
textStream := make(chan string, 3)
@@ -252,7 +252,7 @@ func TestOpenAIService_SynthesizeStream_PartialFailure(t *testing.T) {
})
defer srv.Close()
svc := NewOpenAIService("test-key", "", "alloy", srv.URL, 1.0, 5, zap.NewNop().Sugar())
svc := NewOpenAIService("test-key", "tts-1", "alloy", srv.URL, 1.0, 5, 30, zap.NewNop().Sugar())
textStream := sendSentences("第一句", "第二句", "第三句")
@@ -286,7 +286,7 @@ func TestOpenAIService_SynthesizeStream_CustomVoice(t *testing.T) {
})
defer srv.Close()
svc := NewOpenAIService("test-key", "", "alloy", srv.URL, 1.0, 5, zap.NewNop().Sugar())
svc := NewOpenAIService("test-key", "tts-1", "alloy", srv.URL, 1.0, 5, 30, zap.NewNop().Sugar())
textStream := sendSentences("你好")

84
backend/internal/config/config.go
View File

@@ -11,12 +11,19 @@ import (
// Config 应用配置。
type Config struct {
App AppConfig `mapstructure:"app"`
Server ServerConfig `mapstructure:"server"`
Redis RedisConfig `mapstructure:"redis"`
AI AIConfig `mapstructure:"ai"`
Storage StorageConfig `mapstructure:"storage"`
Log LogConfig `mapstructure:"log"`
App AppConfig `mapstructure:"app"`
Server ServerConfig `mapstructure:"server"`
Session SessionConfig `mapstructure:"session"`
Redis RedisConfig `mapstructure:"redis"`
AI AIConfig `mapstructure:"ai"`
Storage StorageConfig `mapstructure:"storage"`
Log LogConfig `mapstructure:"log"`
}
// SessionConfig 会话管理配置。
type SessionConfig struct {
TTL int `mapstructure:"ttl"` // 会话过期时间（分钟）
MaxHistory int `mapstructure:"max_history"` // 对话历史上限（条）
}
type AppConfig struct {
@@ -25,10 +32,14 @@ type AppConfig struct {
}
type ServerConfig struct {
Host string `mapstructure:"host"`
Port int `mapstructure:"port"`
ReadTimeout int `mapstructure:"read_timeout"`
WriteTimeout int `mapstructure:"write_timeout"`
Host string `mapstructure:"host"`
Port int `mapstructure:"port"`
ReadTimeout int `mapstructure:"read_timeout"`
WriteTimeout int `mapstructure:"write_timeout"`
HeartbeatInterval int `mapstructure:"heartbeat_interval"` // 心跳检查间隔（秒）
HeartbeatTimeout int `mapstructure:"heartbeat_timeout"` // 心跳超时（秒）
ShutdownTimeout int `mapstructure:"shutdown_timeout"` // 优雅关闭超时（秒）
AllowedOrigins []string `mapstructure:"allowed_origins"` // CORS 允许的来源，空表示允许所有
}
// Addr 返回 host:port 地址。
@@ -49,28 +60,34 @@ type AIConfig struct {
}
type STTConfig struct {
Provider string `mapstructure:"provider"`
APIKey string `mapstructure:"api_key"`
Model string `mapstructure:"model"`
Endpoint string `mapstructure:"endpoint"`
Provider string `mapstructure:"provider"`
APIKey string `mapstructure:"api_key"`
Model string `mapstructure:"model"`
Endpoint string `mapstructure:"endpoint"`
Timeout int `mapstructure:"timeout"` // STT 超时（秒）
HTTPClientTimeout int `mapstructure:"http_client_timeout"` // HTTP 客户端超时（秒）
}
type LLMConfig struct {
Provider string `mapstructure:"provider"`
APIKey string `mapstructure:"api_key"`
Model string `mapstructure:"model"`
Endpoint string `mapstructure:"endpoint"`
Timeout int `mapstructure:"timeout"`
Provider string `mapstructure:"provider"`
APIKey string `mapstructure:"api_key"`
Model string `mapstructure:"model"`
Endpoint string `mapstructure:"endpoint"`
Timeout int `mapstructure:"timeout"`
HTTPClientTimeout int `mapstructure:"http_client_timeout"` // HTTP 客户端超时（秒）
}
type TTSConfig struct {
Provider string `mapstructure:"provider"`
APIKey string `mapstructure:"api_key"`
Model string `mapstructure:"model"`
Voice string `mapstructure:"voice"`
Speed float64 `mapstructure:"speed"`
Endpoint string `mapstructure:"endpoint"`
Timeout int `mapstructure:"timeout"`
Provider string `mapstructure:"provider"`
APIKey string `mapstructure:"api_key"`
Model string `mapstructure:"model"`
Voice string `mapstructure:"voice"`
Speed float64 `mapstructure:"speed"`
Endpoint string `mapstructure:"endpoint"`
Timeout int `mapstructure:"timeout"`
HTTPClientTimeout int `mapstructure:"http_client_timeout"` // HTTP 客户端超时（秒）
OutputFormat string `mapstructure:"output_format"` // 输出格式：mp3/wav
SampleRate int `mapstructure:"sample_rate"` // 输出采样率
}
type StorageConfig struct {
@@ -91,28 +108,42 @@ func Load() (*Config, error) {
v.AddConfigPath(".")
v.AddConfigPath("./config")
v.AddConfigPath("./backend")
v.AddConfigPath("..") // 兼容从 backend/cmd/ 启动
v.AddConfigPath("../..") // 兼容从 backend/cmd/server/ 启动
// 默认值
v.SetDefault("app.env", "dev")
v.SetDefault("app.version", "dev")
v.SetDefault("server.host", "0.0.0.0")
v.SetDefault("server.port", 8080)
v.SetDefault("server.read_timeout", 30)
v.SetDefault("server.write_timeout", 30)
v.SetDefault("server.heartbeat_interval", 30)
v.SetDefault("server.heartbeat_timeout", 60)
v.SetDefault("server.shutdown_timeout", 10)
v.SetDefault("session.ttl", 30)
v.SetDefault("session.max_history", 20)
v.SetDefault("redis.addr", "localhost:6379")
v.SetDefault("redis.db", 0)
v.SetDefault("ai.stt.provider", "deepgram")
v.SetDefault("ai.stt.model", "nova-2")
v.SetDefault("ai.stt.endpoint", "wss://api.deepgram.com/v1/listen")
v.SetDefault("ai.stt.timeout", 5)
v.SetDefault("ai.stt.http_client_timeout", 30)
v.SetDefault("ai.llm.provider", "openai")
v.SetDefault("ai.llm.model", "gpt-4o")
v.SetDefault("ai.llm.endpoint", "https://api.openai.com/v1")
v.SetDefault("ai.llm.timeout", 10)
v.SetDefault("ai.llm.http_client_timeout", 60)
v.SetDefault("ai.tts.provider", "openai")
v.SetDefault("ai.tts.model", "tts-1")
v.SetDefault("ai.tts.voice", "mimo_default")
v.SetDefault("ai.tts.speed", 1.0)
v.SetDefault("ai.tts.endpoint", "https://api.openai.com/v1")
v.SetDefault("ai.tts.timeout", 5)
v.SetDefault("ai.tts.http_client_timeout", 30)
v.SetDefault("ai.tts.output_format", "mp3")
v.SetDefault("ai.tts.sample_rate", 24000)
v.SetDefault("storage.driver", "memory")
v.SetDefault("log.level", "info")
v.SetDefault("log.format", "console")
@@ -134,6 +165,7 @@ func Load() (*Config, error) {
// 按优先级尝试：当前目录、上级目录（兼容从 backend/ 或项目根目录启动）
_ = godotenv.Load()
_ = godotenv.Load("../.env")
_ = godotenv.Load("../../.env") // 兼容从 backend/cmd/server/ 启动
// 环境变量覆盖
v.SetEnvPrefix("CAMTALK")

41
backend/internal/orchestrator/pipeline.go
View File

@@ -11,6 +11,7 @@ import (
"github.com/hhs/camtalk/internal/ai/llm"
"github.com/hhs/camtalk/internal/ai/stt"
"github.com/hhs/camtalk/internal/ai/tts"
"github.com/hhs/camtalk/internal/config"
"github.com/hhs/camtalk/internal/logger"
"github.com/hhs/camtalk/internal/models"
"github.com/hhs/camtalk/internal/session"
@@ -18,13 +19,15 @@ import (
// Pipeline 实现 Orchestrator 接口，管理 STT → LLM → TTS 流式管道。
type Pipeline struct {
sttService stt.Service
llmService llm.Service
ttsService tts.Service
sessionMgr session.Manager
model string // LLM 模型名，用于 llm_done 上报
ttsVoice string // TTS 音色
ttsSpeed float64 // TTS 语速
sttService stt.Service
llmService llm.Service
ttsService tts.Service
sessionMgr session.Manager
model string // LLM 模型名，用于 llm_done 上报
ttsVoice string // TTS 音色
ttsSpeed float64 // TTS 语速
ttsOutputFmt string // TTS 输出格式
ttsSampleRate int // TTS 输出采样率
}
// New 创建 Pipeline 实例。
@@ -33,18 +36,18 @@ func New(
llmService llm.Service,
ttsService tts.Service,
sessionMgr session.Manager,
model string,
ttsVoice string,
ttsSpeed float64,
cfg *config.Config,
) *Pipeline {
return &Pipeline{
sttService: sttService,
llmService: llmService,
ttsService: ttsService,
sessionMgr: sessionMgr,
model: model,
ttsVoice: ttsVoice,
ttsSpeed: ttsSpeed,
sttService: sttService,
llmService: llmService,
ttsService: ttsService,
sessionMgr: sessionMgr,
model: cfg.AI.LLM.Model,
ttsVoice: cfg.AI.TTS.Voice,
ttsSpeed: cfg.AI.TTS.Speed,
ttsOutputFmt: cfg.AI.TTS.OutputFormat,
ttsSampleRate: cfg.AI.TTS.SampleRate,
}
}
@@ -314,8 +317,8 @@ func (p *Pipeline) synthesizeTTS(
ttsStream, err := p.ttsService.SynthesizeStream(ctx, sentenceCh, tts.Options{
Voice: p.ttsVoice,
Speed: p.ttsSpeed,
OutputFmt: "mp3",
SampleRate: 24000,
OutputFmt: p.ttsOutputFmt,
SampleRate: p.ttsSampleRate,
})
if err != nil {
log.Errorw("TTS 合成启动失败", "error", err)

57
backend/internal/orchestrator/pipeline_test.go
View File

@@ -13,6 +13,7 @@ import (
"github.com/hhs/camtalk/internal/ai/llm"
"github.com/hhs/camtalk/internal/ai/stt"
"github.com/hhs/camtalk/internal/ai/tts"
"github.com/hhs/camtalk/internal/config"
"github.com/hhs/camtalk/internal/logger"
"github.com/hhs/camtalk/internal/models"
)
@@ -254,7 +255,12 @@ func TestProcessQuery_Success(t *testing.T) {
mockSender.On("SendTTSAudio", mock.Anything).Return(nil)
// 创建 Pipeline
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, "gpt-4o", "alloy", 1.0)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, &config.Config{
AI: config.AIConfig{
LLM: config.LLMConfig{Model: "gpt-4o"},
TTS: config.TTSConfig{Voice: "alloy", Speed: 1.0, OutputFormat: "mp3", SampleRate: 24000},
},
})
// 执行
ctx := context.Background()
@@ -305,7 +311,12 @@ func TestProcessQuery_STTError(t *testing.T) {
mockSender.On("SendError", mock.Anything).Return(nil)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, "gpt-4o", "alloy", 1.0)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, &config.Config{
AI: config.AIConfig{
LLM: config.LLMConfig{Model: "gpt-4o"},
TTS: config.TTSConfig{Voice: "alloy", Speed: 1.0, OutputFormat: "mp3", SampleRate: 24000},
},
})
ctx := context.Background()
err := pipeline.ProcessQuery(ctx, "session-123", req, nil, mockSender)
@@ -357,7 +368,12 @@ func TestProcessQuery_LLMError(t *testing.T) {
mockSender.On("SendError", mock.Anything).Return(nil)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, "gpt-4o", "alloy", 1.0)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, &config.Config{
AI: config.AIConfig{
LLM: config.LLMConfig{Model: "gpt-4o"},
TTS: config.TTSConfig{Voice: "alloy", Speed: 1.0, OutputFormat: "mp3", SampleRate: 24000},
},
})
ctx := context.Background()
err := pipeline.ProcessQuery(ctx, "session-123", req, nil, mockSender)
@@ -415,7 +431,12 @@ func TestProcessQuery_TTSError(t *testing.T) {
mockTTS.On("SynthesizeStream", mock.Anything, mock.Anything, mock.Anything).
Return(nil, errors.New("TTS service unavailable"))
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, "gpt-4o", "alloy", 1.0)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, &config.Config{
AI: config.AIConfig{
LLM: config.LLMConfig{Model: "gpt-4o"},
TTS: config.TTSConfig{Voice: "alloy", Speed: 1.0, OutputFormat: "mp3", SampleRate: 24000},
},
})
ctx := context.Background()
err := pipeline.ProcessQuery(ctx, "session-123", req, nil, mockSender)
@@ -485,7 +506,12 @@ func TestProcessQuery_ContextCancelled(t *testing.T) {
}()
mockTTS.On("SynthesizeStream", mock.Anything, mock.Anything, mock.Anything).Return((<-chan tts.Chunk)(ttsCh), nil)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, "gpt-4o", "alloy", 1.0)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, &config.Config{
AI: config.AIConfig{
LLM: config.LLMConfig{Model: "gpt-4o"},
TTS: config.TTSConfig{Voice: "alloy", Speed: 1.0, OutputFormat: "mp3", SampleRate: 24000},
},
})
// 创建可取消的上下文
ctx, cancel := context.WithCancel(context.Background())
@@ -545,7 +571,12 @@ func TestProcessQuery_DisabledTTS(t *testing.T) {
mockSender.On("SendLLMChunk", mock.Anything).Return(nil)
mockSender.On("SendLLMDone", mock.Anything).Return(nil)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, "gpt-4o", "alloy", 1.0)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, &config.Config{
AI: config.AIConfig{
LLM: config.LLMConfig{Model: "gpt-4o"},
TTS: config.TTSConfig{Voice: "alloy", Speed: 1.0, OutputFormat: "mp3", SampleRate: 24000},
},
})
ctx := context.Background()
err := pipeline.ProcessQuery(ctx, "session-123", req, nil, mockSender)
@@ -617,7 +648,12 @@ func TestProcessQuery_InvalidAudio(t *testing.T) {
mockSender.On("SendError", mock.Anything).Return(nil)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, "gpt-4o", "alloy", 1.0)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, &config.Config{
AI: config.AIConfig{
LLM: config.LLMConfig{Model: "gpt-4o"},
TTS: config.TTSConfig{Voice: "alloy", Speed: 1.0, OutputFormat: "mp3", SampleRate: 24000},
},
})
ctx := context.Background()
err := pipeline.ProcessQuery(ctx, "session-123", req, nil, mockSender)
@@ -650,7 +686,12 @@ func TestProcessQuery_SessionNotFound(t *testing.T) {
mockSender.On("SendError", mock.Anything).Return(nil)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, "gpt-4o", "alloy", 1.0)
pipeline := New(mockSTT, mockLLM, mockTTS, mockSession, &config.Config{
AI: config.AIConfig{
LLM: config.LLMConfig{Model: "gpt-4o"},
TTS: config.TTSConfig{Voice: "alloy", Speed: 1.0, OutputFormat: "mp3", SampleRate: 24000},
},
})
ctx := context.Background()
err := pipeline.ProcessQuery(ctx, "session-123", req, nil, mockSender)

42
backend/internal/ws/handler.go
View File

@@ -10,6 +10,7 @@ import (
"github.com/gin-gonic/gin"
"github.com/gorilla/websocket"
"github.com/hhs/camtalk/internal/config"
"github.com/hhs/camtalk/internal/errors"
"github.com/hhs/camtalk/internal/logger"
"github.com/hhs/camtalk/internal/models"
@@ -17,8 +18,23 @@ import (
"github.com/hhs/camtalk/internal/session"
)
var upgrader = websocket.Upgrader{
CheckOrigin: func(r *http.Request) bool { return true }, // 开发阶段允许所有来源
// newUpgrader 根据配置创建 WebSocket upgrader。
func newUpgrader(cfg *config.Config) websocket.Upgrader {
allowedOrigins := cfg.Server.AllowedOrigins
return websocket.Upgrader{
CheckOrigin: func(r *http.Request) bool {
if len(allowedOrigins) == 0 {
return true // 未配置则允许所有来源（开发模式）
}
origin := r.Header.Get("Origin")
for _, o := range allowedOrigins {
if o == origin || o == "*" {
return true
}
}
return false
},
}
}
// Client 代表一个 WebSocket 客户端连接。
@@ -75,13 +91,21 @@ func (w *WSClient) SendError(err models.WsError) error {
}
// ServeWS 处理 WebSocket 升级请求。
func ServeWS(sessionMgr session.Manager, orch orchestrator.Orchestrator) gin.HandlerFunc {
func ServeWS(sessionMgr session.Manager, orch orchestrator.Orchestrator, cfg *config.Config) gin.HandlerFunc {
upgrader := newUpgrader(cfg)
heartbeatInterval := time.Duration(cfg.Server.HeartbeatInterval) * time.Second
heartbeatTimeout := time.Duration(cfg.Server.HeartbeatTimeout) * time.Second
version := cfg.App.Version
maxHistory := cfg.Session.MaxHistory
return func(c *gin.Context) {
serveWS(c, sessionMgr, orch)
serveWS(c, sessionMgr, orch, upgrader, heartbeatInterval, heartbeatTimeout, version, maxHistory)
}
}
func serveWS(c *gin.Context, sessionMgr session.Manager, orch orchestrator.Orchestrator) {
func serveWS(c *gin.Context, sessionMgr session.Manager, orch orchestrator.Orchestrator,
upgrader websocket.Upgrader, heartbeatInterval, heartbeatTimeout time.Duration, version string, maxHistory int) {
conn, err := upgrader.Upgrade(c.Writer, c.Request, nil)
if err != nil {
logger.Log.Errorw("websocket upgrade failed", "error", err)
@@ -108,7 +132,7 @@ func serveWS(c *gin.Context, sessionMgr session.Manager, orch orchestrator.Orche
_ = client.SendJSON(models.WsConnected{
Type: "connected",
SessionID: sessionID,
ServerVersion: "0.1.0",
ServerVersion: version,
})
logger.Log.Infow("client connected", "session", sessionID)
@@ -122,12 +146,12 @@ func serveWS(c *gin.Context, sessionMgr session.Manager, orch orchestrator.Orche
// 启动心跳检查 goroutine
done := make(chan struct{})
go func() {
ticker := time.NewTicker(30 * time.Second)
ticker := time.NewTicker(heartbeatInterval)
defer ticker.Stop()
for {
select {
case <-ticker.C:
if time.Since(lastPong) > 60*time.Second {
if time.Since(lastPong) > heartbeatTimeout {
logger.Log.Warnw("heartbeat timeout", "session", sessionID)
conn.Close()
return
@@ -181,7 +205,7 @@ func serveWS(c *gin.Context, sessionMgr session.Manager, orch orchestrator.Orche
}
// 获取对话历史
history, _ := client.sessionMgr.GetHistory(context.Background(), sessionID, 20)
history, _ := client.sessionMgr.GetHistory(context.Background(), sessionID, maxHistory)
// 创建可取消的 context
ctx, cancel := context.WithCancel(context.Background())

10
backend/internal/ws/handler_test.go
View File

@@ -13,6 +13,7 @@ import (
"github.com/stretchr/testify/require"
"context"
"github.com/hhs/camtalk/internal/config"
"github.com/hhs/camtalk/internal/logger"
"github.com/hhs/camtalk/internal/models"
"github.com/hhs/camtalk/internal/orchestrator"
@@ -138,7 +139,12 @@ func setupTestServer(t *testing.T, orch orchestrator.Orchestrator) (*httptest.Se
t.Cleanup(func() { sessionMgr.Stop() })
r := gin.New()
r.GET("/ws", ServeWS(sessionMgr, orch))
cfg := &config.Config{
App: config.AppConfig{Version: "test"},
Server: config.ServerConfig{HeartbeatInterval: 30, HeartbeatTimeout: 60},
Session: config.SessionConfig{MaxHistory: 20},
}
r.GET("/ws", ServeWS(sessionMgr, orch, cfg))
srv := httptest.NewServer(r)
@@ -181,7 +187,7 @@ func TestWS_Connected(t *testing.T) {
msg := readJSON(t, conn)
assert.Equal(t, "connected", msg["type"])
assert.NotEmpty(t, msg["session_id"])
assert.Equal(t, "0.1.0", msg["server_version"])
assert.Equal(t, "test", msg["server_version"])
}
// TestWS_PingPong 验证 ping/pong 心跳。

5
frontend/src/lib/websocket.ts
View File

@@ -6,7 +6,10 @@
import type { ClientMessage, ServerMessage, WsMessage } from "../types";
const WS_URL = "ws://localhost:8080/ws";
// WebSocket 地址：优先使用环境变量，否则基于当前页面地址自动推导
const WS_URL =
import.meta.env.VITE_WS_URL ||
`${window.location.protocol === "https:" ? "wss:" : "ws:"}//${window.location.host}/ws`;
const PING_INTERVAL = 30_000; // 30 秒心跳
const MAX_RECONNECT_DELAY = 30_000; // 最大重连延迟 30 秒

5
功能创意.md
View File

@@ -4,3 +4,8 @@
4.手动对话功能
5.视频框大小可调整，可最小化然后拖动
6.增加对话情景选择功能，参考豆包（面试官，英语老师，辩论赛，同声翻译等）
修复bug:
对话语音重复
麦克风开关