Conversation
Voice Assignments
Default voice
No persona templates available. Please check that templates are properly configured in the manifest.
Voice Profiles
Loading profiles...
TTS Provider Voices
ElevenLabs:
Gemini:
Orpheus:
PlayAI v3:
DIA:
F5:
Kokoro:
Loading Voice Profiles...
Please wait while voice profiles are loaded from configuration.
Voice Profiles Could Not Be Loaded
Unable to load voice profiles from data/voice-profiles.json. Please ensure:
- The application is served via HTTP (not file://)
- The voice-profiles.json file exists in the data directory
Current Template Voice Assignments
Voice assignments for template:
Speaker
All data is stored locally in your browser. Nothing is sent to the cloud except API calls to AI providers.
API Keys
Get your API key from OpenAI Platform
Get your API key from Google AI Studio
Get your API key from xAI Console
Get your API key from DeepSeek Platform
Get your API key from ElevenLabs API
Get your API key from fal.ai dashboard
Text Generation
Text-to-Speech
Text-to-Speech: Each message gets its own audio file with individual play buttons.
Dialogue: Entire conversation is generated as a single audio file optimized for natural flow.
Dialogue: Entire conversation is generated as a single audio file optimized for natural flow.
Orpheus: State-of-the-art empathetic speech synthesis
PlayAI V3: Advanced voice cloning with emotion support
PlayAI Dialog: Multi-speaker dialogue generation
Dia TTS: Natural speech synthesis with realistic nonverbals like laughter, breathing, and emotional expressions
Dia TTS Clone: Voice cloning that preserves natural speech patterns including nonverbals and emotions
F5-TTS: High-quality voice synthesis and cloning
Kokoro: American English voice synthesis
Chatterbox: Fast and efficient TTS
Temperature Control
Lower values (0.0-0.5) make outputs more focused and deterministic. Higher values (0.5-2.0) make outputs more creative and varied.
Temperature Control
⚠️ The selected model () is a reasoning model that doesn't support temperature control. Temperature is fixed at 1.0.
Request
URL:
Method:
Type:
Endpoint:
Voice ID:
Request ID:
Attempt:
Headers
Body
Response
URL:
Status:
Duration:
Type:
Response Size:
Audio Size:
Audio Type:
Voice ID:
Request ID:
Attempt:
Input Count:
Speaker Count:
Dialogue Length:
Candidates Count:
Headers
Data
Error
Error
Voice ID:
Type:
Error Message
Stack Trace
Request (Legacy)
Response (Legacy)
Error (Legacy)
Duration (Legacy)
No debug entries yet. Enable debug logging and make some API calls to see them here.
Voice Model Testing
Enter a prompt to generate dialogue using each TTS model's specific prompting guide. Compare how different models interpret and generate content.
Voice Models to Test
of models selected
Total Models:
Completed:
Failed:
Duration:
Error:
Text:
Audio:
Generation Failed
No Audio Generated