Add Voice Capabilities to Nevron (ElevenLabs Integration)

[gromdimon]

Description

We need to add voice capabilities to Nevron by integrating ElevenLabs for text-to-speech (TTS) generation. This will allow the agent to respond with audio, making interactions more engaging and accessible.

The feature should:

1. Convert agent responses into speech using ElevenLabs’ TTS API.

---

Requirements

1. ElevenLabs API Integration

• Use ElevenLabs TTS API for generating voice responses.
• Support different voices and accents provided by ElevenLabs.
• Ensure API keys and settings are configurable via .env.
• Add this to new Executor

2. Voice Configuration

• Allow users to choose different voices (e.g., Rachel, Adam, etc.).
• Support custom speed, pitch, and tone settings.
• Ensure API limits are handled properly.

3. Testing

• Validate the API response times and audio quality.
• Ensure audio files are stored securely and cleared after use.