feat(voice/tts): add full voice profile to TTSConfig

Expand TTSConfig with model, format, and speed so the agent's voice is
fully configurable from config.json without touching code.

- TTSConfig gains: Model, Format, Speed fields (all env-overridable)
- KokoroSynthesizer: add TTSProfile struct + NewKokoroSynthesizerFromProfile()
  NewKokoroSynthesizer() is kept as a convenience wrapper (backward-compat)
- kokoroRequest: pass format and speed through to the API
- Temp file extension follows configured format (mp3/wav/ogg/etc.)
- main.go: wire all profile fields from config
- config.example.json: updated with model/format/speed examples

Full profile example:
  "tts": {
    "enabled": true,
    "api_base": "http://localhost:8100",
    "voice":   "af_nova",
    "model":   "kokoro",
    "format":  "mp3",
    "speed":   1.0
  }

cmd/picoclaw/main.go

@@ -636,7 +636,13 @@ func gatewayCmd() {
 
 	// Attach TTS synthesis callbacks to the message tool (enables voice=true).
 	if cfg.Tools.TTS.Enabled {
-		synthesizer := voice.NewKokoroSynthesizer(cfg.Tools.TTS.APIBase, cfg.Tools.TTS.Voice)
+		synthesizer := voice.NewKokoroSynthesizerFromProfile(voice.TTSProfile{
+			APIBase: cfg.Tools.TTS.APIBase,
+			Voice:   cfg.Tools.TTS.Voice,
+			Model:   cfg.Tools.TTS.Model,
+			Format:  cfg.Tools.TTS.Format,
+			Speed:   cfg.Tools.TTS.Speed,
+		})
 		if synthesizer.IsAvailable() {
 			logger.InfoCF("voice", "TTS enabled — voice=true supported in message tool", map[string]interface{}{
 				"api_base": cfg.Tools.TTS.APIBase,

config/config.example.json

@@ -14,7 +14,9 @@
       "enabled": false,
       "token": "YOUR_TELEGRAM_BOT_TOKEN",
       "proxy": "",
-      "allow_from": ["YOUR_USER_ID"]
+      "allow_from": [
+        "YOUR_USER_ID"
+      ]
     },
     "discord": {
       "enabled": false,
@@ -119,6 +121,14 @@
         "api_key": "YOUR_BRAVE_API_KEY",
         "max_results": 5
       }
+    },
+    "tts": {
+      "enabled": false,
+      "api_base": "http://localhost:8100",
+      "voice": "en_us-lessac-medium",
+      "model": "tts-1",
+      "format": "mp3",
+      "speed": 1.0
     }
   },
   "heartbeat": {

pkg/config/config.go

@@ -217,9 +217,12 @@ type WhisperConfig struct {
 }
 
 type TTSConfig struct {
-	Enabled bool   `json:"enabled" env:"PICOCLAW_TOOLS_TTS_ENABLED"`
+	Enabled bool    `json:"enabled" env:"PICOCLAW_TOOLS_TTS_ENABLED"`
-	APIBase string `json:"api_base" env:"PICOCLAW_TOOLS_TTS_API_BASE"`
+	APIBase string  `json:"api_base" env:"PICOCLAW_TOOLS_TTS_API_BASE"`
-	Voice   string `json:"voice" env:"PICOCLAW_TOOLS_TTS_VOICE"`
+	Voice   string  `json:"voice" env:"PICOCLAW_TOOLS_TTS_VOICE"`
+	Model   string  `json:"model" env:"PICOCLAW_TOOLS_TTS_MODEL"`
+	Format  string  `json:"format" env:"PICOCLAW_TOOLS_TTS_FORMAT"`
+	Speed   float64 `json:"speed" env:"PICOCLAW_TOOLS_TTS_SPEED"`
 }
 
 type ToolsConfig struct {
@@ -343,6 +346,9 @@ func DefaultConfig() *Config {
 				Enabled: false,
 				APIBase: "http://localhost:8100",
 				Voice:   "en_us-lessac-medium",
+				Model:   "tts-1",
+				Format:  "mp3",
+				Speed:   1.0,
 			},
 		},
 		Heartbeat: HeartbeatConfig{

pkg/voice/kokoro.go

@@ -13,41 +13,75 @@ import (
 	"github.com/sipeed/picoclaw/pkg/logger"
 )
 
-// KokoroSynthesizer uses a Kokoro TTS server (OpenAI-compatible /v1/audio/speech API).
+// KokoroSynthesizer uses any OpenAI-compatible /v1/audio/speech endpoint
+// (Kokoro, Piper, Chatterbox, OpenAI, etc.).
 type KokoroSynthesizer struct {
 	apiBase    string
 	voice      string
 	model      string
+	format     string
+	speed      float64
 	httpClient *http.Client
 }
 
 type kokoroRequest struct {
-	Model  string `json:"model"`
+	Model  string  `json:"model"`
-	Input  string `json:"input"`
+	Input  string  `json:"input"`
-	Voice  string `json:"voice"`
+	Voice  string  `json:"voice"`
-	Format string `json:"response_format,omitempty"`
+	Format string  `json:"response_format,omitempty"`
+	Speed  float64 `json:"speed,omitempty"`
 }
 
-// NewKokoroSynthesizer creates a Kokoro TTS client.
+// TTSProfile holds the full voice profile for the synthesizer.
-// apiBase defaults to "http://localhost:8102".
+type TTSProfile struct {
-// voice defaults to "af_nova".
+	APIBase string
+	Voice   string
+	Model   string
+	Format  string
+	Speed   float64
+}
+
+// NewKokoroSynthesizer creates a TTS client from a voice profile.
+// Sensible defaults are applied for any zero-value field.
 func NewKokoroSynthesizer(apiBase, voice string) *KokoroSynthesizer {
-	if apiBase == "" {
+	return NewKokoroSynthesizerFromProfile(TTSProfile{
-		apiBase = "http://localhost:8102"
+		APIBase: apiBase,
+		Voice:   voice,
+	})
+}
+
+// NewKokoroSynthesizerFromProfile creates a TTS client with full profile control.
+func NewKokoroSynthesizerFromProfile(p TTSProfile) *KokoroSynthesizer {
+	if p.APIBase == "" {
+		p.APIBase = "http://localhost:8100"
 	}
-	if voice == "" {
+	if p.Voice == "" {
-		voice = "af_nova"
+		p.Voice = "en_us-lessac-medium"
+	}
+	if p.Model == "" {
+		p.Model = "tts-1"
+	}
+	if p.Format == "" {
+		p.Format = "mp3"
+	}
+	if p.Speed == 0 {
+		p.Speed = 1.0
 	}
 
-	logger.InfoCF("voice", "Creating Kokoro TTS synthesizer", map[string]interface{}{
+	logger.InfoCF("voice", "Creating TTS synthesizer", map[string]interface{}{
-		"api_base": apiBase,
+		"api_base": p.APIBase,
-		"voice":    voice,
+		"voice":    p.Voice,
+		"model":    p.Model,
+		"format":   p.Format,
+		"speed":    p.Speed,
 	})
 
 	return &KokoroSynthesizer{
-		apiBase: apiBase,
+		apiBase: p.APIBase,
-		voice:   voice,
+		voice:   p.Voice,
-		model:   "kokoro",
+		model:   p.Model,
+		format:  p.Format,
+		speed:   p.Speed,
 		httpClient: &http.Client{
 			Timeout: 60 * time.Second,
 		},
@@ -66,7 +100,8 @@ func (s *KokoroSynthesizer) Synthesize(ctx context.Context, text string) (string
 		Model:  s.model,
 		Input:  text,
 		Voice:  s.voice,
-		Format: "mp3",
+		Format: s.format,
+		Speed:  s.speed,
 	}
 
 	bodyBytes, err := json.Marshal(reqBody)
@@ -93,7 +128,7 @@ func (s *KokoroSynthesizer) Synthesize(ctx context.Context, text string) (string
 	}
 
 	// Write audio to temp file
-	tmpFile, err := os.CreateTemp("", "picoclaw-tts-*.mp3")
+	tmpFile, err := os.CreateTemp("", "picoclaw-tts-*."+s.format)
 	if err != nil {
 		return "", fmt.Errorf("failed to create temp audio file: %w", err)
 	}