Align Gemini audio defaults with TTS API

3 files changed, 35 insertions, 8 deletions

diff --git a/internal/audio/gemini_provider.go b/internal/audio/gemini_provider.go
index 8d0c517..9900005 100644
--- a/internal/audio/gemini_provider.go
+++ b/internal/audio/gemini_provider.go
@@ -14,7 +14,7 @@ import (
 )
 
 const (
-	defaultGeminiTTSModel  = "gemini-2.5-flash"
+	defaultGeminiTTSModel  = "gemini-2.5-flash-preview-tts"
 	geminiTTSLanguageCode  = "bg"
 	geminiTTSChannels      = 1
 	geminiTTSSampleRate    = 24000
diff --git a/internal/audio/provider.go b/internal/audio/provider.go
index d750f31..06ed6d0 100644
--- a/internal/audio/provider.go
+++ b/internal/audio/provider.go
@@ -32,7 +32,7 @@ type Config struct {
 
 	// Gemini-specific settings
 	GoogleAPIKey   string
-	GeminiTTSModel string  // "gemini-2.5-flash"
+	GeminiTTSModel string  // "gemini-2.5-flash-preview-tts"
 	GeminiVoice    string  // One of GeminiVoices, or empty for the model default.
 	GeminiSpeed    float64 // Prompt hint for desired speech speed
 }
@@ -42,13 +42,13 @@ func DefaultProviderConfig() *Config {
 	return &Config{
 		Provider:     "gemini",
 		OutputDir:    "./",
-		OutputFormat: "mp3",
+		OutputFormat: "wav",
 		OpenAIModel:  "gpt-4o-mini-tts", // New model with voice instructions support
 		OpenAIVoice:  "alloy",
 		OpenAISpeed:  1.0,
 		// OpenAISpeed:       0.98, // Default speed for clarity
 		OpenAIInstruction: "You are speaking Bulgarian language (български език). Pronounce the Bulgarian text with authentic Bulgarian phonetics, not Russian. Speak slowly and clearly for language learners.",
-		GeminiTTSModel:    "gemini-2.5-flash",
+		GeminiTTSModel:    "gemini-2.5-flash-preview-tts",
 		GeminiSpeed:       1.0,
 	}
 }
diff --git a/internal/audio/provider_test.go b/internal/audio/provider_test.go
index 0b04d20..64fbcc7 100644
--- a/internal/audio/provider_test.go
+++ b/internal/audio/provider_test.go
@@ -3,6 +3,8 @@ package audio
 import (
 	"context"
 	"errors"
+	"path/filepath"
+	"strings"
 	"testing"
 )
 
@@ -34,8 +36,8 @@ func TestDefaultProviderConfig(t *testing.T) {
 		t.Errorf("Expected provider 'gemini', got '%s'", config.Provider)
 	}
 
-	if config.OutputFormat != "mp3" {
-		t.Errorf("Expected output format 'mp3', got '%s'", config.OutputFormat)
+	if config.OutputFormat != "wav" {
+		t.Errorf("Expected output format 'wav', got '%s'", config.OutputFormat)
 	}
 
 	if config.OpenAIModel != "gpt-4o-mini-tts" {
@@ -50,8 +52,8 @@ func TestDefaultProviderConfig(t *testing.T) {
 		t.Errorf("Expected OpenAI speed 1.0, got %f", config.OpenAISpeed)
 	}
 
-	if config.GeminiTTSModel != "gemini-2.5-flash" {
-		t.Errorf("Expected Gemini TTS model 'gemini-2.5-flash', got '%s'", config.GeminiTTSModel)
+	if config.GeminiTTSModel != "gemini-2.5-flash-preview-tts" {
+		t.Errorf("Expected Gemini TTS model 'gemini-2.5-flash-preview-tts', got '%s'", config.GeminiTTSModel)
 	}
 
 	if config.GeminiSpeed != 1.0 {
@@ -59,6 +61,31 @@ func TestDefaultProviderConfig(t *testing.T) {
 	}
 }
 
+func TestDefaultProviderConfigIsGeminiCompatible(t *testing.T) {
+	config := DefaultProviderConfig()
+
+	if config.Provider != "gemini" {
+		t.Fatalf("DefaultProviderConfig() Provider = %q, want %q", config.Provider, "gemini")
+	}
+
+	if config.GeminiTTSModel != defaultGeminiTTSModel {
+		t.Fatalf("DefaultProviderConfig() GeminiTTSModel = %q, want %q", config.GeminiTTSModel, defaultGeminiTTSModel)
+	}
+
+	outputFile := filepath.Join(t.TempDir(), "audio."+config.OutputFormat)
+	if filepath.Ext(outputFile) != ".wav" {
+		t.Fatalf("DefaultProviderConfig() output file %q is incompatible with Gemini TTS", outputFile)
+	}
+
+	if !strings.HasSuffix(config.GeminiTTSModel, "-tts") {
+		t.Fatalf("DefaultProviderConfig() GeminiTTSModel = %q, want a TTS model variant", config.GeminiTTSModel)
+	}
+
+	if err := writeGeminiAudioFile(outputFile, []byte{0x11, 0x22}, "audio/pcm"); err != nil {
+		t.Fatalf("writeGeminiAudioFile() with default Gemini output failed: %v", err)
+	}
+}
+
 func TestNewProvider(t *testing.T) {
 	tests := []struct {
 		name         string