Seamless AI Voice Integration

I have a guided-relaxation track in WAV format that already contains continuous background music, the Ukrainian instructor’s original speech, and an English translation that I generated with Google’s AI voice. What I now need is a single, polished master where the music bed flows without a hitch, no matter which voice is speaking.

Here is the work in a nutshell:
• Keep the Ukrainian teacher’s voice exactly as it is.
• Replace the current English translator track with my supplied AI voice recording (same script, a little longer is fine).
• Maintain identical overall loudness to the original—no boosts or dips—so the listener never notices a hand-off between speaker and translation.
• Ensure the background music remains seamless beneath both voices, with no audible edits, clicks, or timing issues.

Deliverables
1. Final mixed relaxation track in WAV (and MP3 for convenience).
2. A stem or project file showing separate music, Ukrainian, and AI-English tracks, so small tweaks can be made later if needed.

Acceptance criteria
• Smooth, gap-free music throughout.
• Voices balanced at the same perceived level.
• No artifacts, pops, or abrupt fades when switching speakers.

Feel free to use any DAW or mastering tools you prefer—Reaper, Pro Tools, Audition, Logic—so long as the outcome meets the quality above.