Cinematic Audio Producer (AI Voices + RVC + Emotional FX + Sound Design) Long-Term

We are producing cinematic, multi-character dark-romance audiobooks for a premium entertainment platform.
We are hiring one specialist OR a small team who can create high-quality, emotionally immersive audio using a combination of:

AI TTS

RVC (Real-Time Voice Conversion)

Emotional reaction FX

Cinematic ambience & SFX

Professional mixing


This role requires no human recording.
All voices must be created using AI + RVC.

This job is long-term, with guaranteed monthly work.

MANDATORY TOOLSET
1. One Clean TTS Tool (for generating all base dialogue)

You must use ONE of the following:

Fish Audio

Revoicer

ElevenLabs (allowed, but be aware of line restrictions)

or any similar clean TTS tool that outputs non-distorted speech

2. RVC (Real-Time Voice Conversion v2)

Used to convert all base TTS lines into:

the correct character voice

the correct tone

the correct emotional delivery

consistent voice identity across all chapters

3. Audio Editing Software

Use one of:

Adobe Audition

Audacity


For trimming, noise removal, preparing dialogue.

4. Mixing Software

Use one of:

DaVinci Resolve

Adobe Audition

Adobe Premiere Pro


For full cinematic mixing.

5. SFX Library

Use Soundly or your own high-quality SFX library for:

ambience (subtle)

environment

footsteps

fabric/bed movement

keys, doors

atmosphere layers

VOICE MODELS REQUIREMENT

Character Voice Models (RVC)

We provide:

character descriptions

tone

emotion style

You must create or select RVC models that match each character.

Once created, these voices are reused across many books. Once we have most of the cast like multiple male voices and female voice we eventually wont need to make new ones anymore. Probably after some audiobooks. As we have around 48+ books for the upcoming year that need to be made into audiobooks.

Reaction Voice Models (RVC)

For emotional realism, you must use reaction-focused RVC models to convert safe reaction FX such as:

breaths

gasps

sighs

soft vocal reactions

strained / tension-based vocalizations

emotional vocal textures


These must be layered under dialogue for cinematic performance.

BEST WORKFLOW FOR REACTIONS (MANDATORY)

To keep production fast and realistic, you must

1. Take  emotional FX (breaths, gasps, sighs, tension sounds, emotional vocalizations) from a high-quality FX library


2. Run them through the character’s RVC model


3. Layer them under the dialogue during mixing



This produces:

Character-style breaths,emotional gasps, tension exhales etc
All matching the character’s voice identity.

MANDATORY HARDWARE

Windows PC

NVIDIA GPU (RTX 2060 minimum)

Good monitoring headphones

Stable internet

WORKFLOW (Step-by-Step)

1. Base Line Generation (TTS)

Generate each line of dialogue from the script.

2. Convert to Character Voice (RVC)

Every line must be processed through the correct character model.

3. Add Emotional FX (RVC Reaction Layer)

Use safe emotional FX → convert in RVC → layer under main dialogue.

4. Cinematic SFX + Ambience

Add room tone, subtle ambience, movement, object interactions.

5. Mix & Master

Blend all layers
Add EQ, reverb, compression
Export polished chapter audio.

6. Deliver Chapter-by-Chapter

Not all at once.
Ensures perfect quality control.

TIMELINE REQUIREMENT

Our production pace requires a team or a solo freelancer with colleagues.

You must deliver:
2 complete audiobooks every 10 days

Each audiobook:

6–7 hours

40–90 chapters

40k–70k words

fully parsed and ready for production


Monthly pace = 4 audiobooks (minimum 4 guaranteed).


BUDGET

$400 per completed audiobook

Paid per approved audiobook or chapter batch milestone. Whatever you prefer.

Guaranteed Minimum: 4audiobooks/month = $1600

(Teams who deliver faster can earn more.)

Budget Increase After 3 Months

After three months of consistent, high-quality delivery,
rate per audiobook will be increased significantly as part of long-term collaboration.

NDA REQUIREMENT

Before receiving any:

scripts

character references

voice models

production assets


ALL members of your team must sign an NDA.
All materials are confidential and IP of our company.

TO APPLY, PLEASE SEND:

1. Audio samples showing AI TTS → RVC conversion


2. Proof of your mixing ability (cinematic style)


3. Description of your workflow


4. Confirmation you use ALL mandatory tools


5. Hardware specs (GPU required)


6. Your availability per month


7. Whether you are a team or solo with colleagues


8. Confirmation you accept $400/book for first 4 audiobooks + guaranteed monthly work