
The Engine Room · Audio
Every voice. Every note. Free first, and cleared to use.
14 audio engines are listed for voiceovers, podcasts, music beds, phone greetings and sound effects. Free engines go first; premium voices ask first.
demo businessWhat it is
The sound behind everything you post
Every voice in your videos, every music bed under them, every greeting your callers hear and every sound effect comes from an audio engine. The OS can call 14 of them. Each one is good at 1 thing, and the router picks the cheapest one that does the job well.
Most runs free. Qwen3-TTS: multilingual voiceovers. Kokoro-82M (planned): fast, clean narration. Dia: podcast dialogue with more than 1 voice. ACE-Step and YuE: instrumentals. ElevenLabs adds feeling when needed, asking first.
2 rules never bend. A cloned voice is a likeness: cloning needs a signed release on file, enforced in code. Every take is mastered (right volume) for video, podcast or phone.
The full list
All 14 audio engines
Each engine carries a license badge, speed and quality rating, cost label and lane. Product fact.
- WhispervoiceFree engine · $0
- Qwen3-TTSvoiceFree engine · $0
- Kokoro-82MvoiceFree engine · $0
- ChatterboxvoiceFree engine · $0
- PipervoiceFree engine · $0
- NeuTTSvoiceFree engine · $0
- DiavoiceFree engine · $0
- ElevenLabs v3voiceMetered · asks first
- ACE-StepvoiceFree engine · $0
- YuEvoiceFree engine · $0
- Stable Audio OpenvoiceFree engine · $0
- Suno v5.5voiceOwned seat
- FreesoundvoiceFree engine · $0
- LoudnormvoiceFree engine · $0
- RNNoisevoiceFree engine · $0
15 in this group, from the Engine Room manifest. Lanes: free engines first ($0), owned seats, metered engines that ask first. Product fact
Stand-out picks
Who does what in audio
The free lane does most work. The paid lane covers the few jobs free can't.
- Qwen3-TTSThe main voiceover engine, in several languagesFree engine · $0
- Kokoro-82M (planned)Fast, clean narration and phone menusLocal · $0
- ChatterboxExpressive voices, cloning only with consentFree engine · $0
- DiaPodcast dialogue with more than 1 speakerFree engine · $0
- Piper (planned)Light voice tuned for phone lines (8 kHz)Local · $0
- NeuTTSThe phone receptionist's voiceFree engine · $0
- ACE-Step / YuEFree instrumental music and lyrics-to-songFree engine · $0
- Stable Audio OpenMusic and sound effectsFree engine · $0
- FreesoundCC0 / CC-BY sound effects, credit trackedFree
- ElevenLabs v3Emotive premium voiceover, voice clone, effectsMetered · asks first
- Suno v5.5Music with vocals and stems, rights flaggedSubscription
- Loudnorm · RNNoiseLoudness mastering and noise removal, always freeLocal · $0
Free engines first. Premium engines only when you approve, with the cost shown before anything runs.
Job by job
The right audio engine for each job
Free first. A paid engine steps in only when needed, and only after you say yes.
| Job | Free first | When it steps up | The rule |
|---|---|---|---|
| Voiceover | Qwen3-TTS, Kokoro-82M | ElevenLabs v3, when the free voices won't carry the read | Mastered to −16 LUFS, broadcast volume |
| Podcast | Dia, multi-voice dialogue | Stays free | −16 LUFS mono; −19 for EU spoken word |
| Music bed | ACE-Step, YuE, Stable Audio Open | Suno, for songs with vocals | Rights flagged before it ships; sits about 15% under the voice |
| Phone greeting and on-hold | Piper (planned) at 8 kHz, NeuTTS | ElevenLabs voices on the line | Mastered for phone audio |
| Sound effects | Freesound library, Stable Audio Open | Stays free | CC0 or CC-BY only, with the credit tracked |
| Clean-up | RNNoise, loudnorm | Stays free | Noise removed, loudness set, every time |
| Voice clone | Chatterbox | ElevenLabs v3 | Signed release on file before a single word |
| Transcripts | Whisper | Stays free | Speech to text, on our free engines |
Product fact: engines and rules from the Engine Room and the Scoring Stage. LUFS is how loud audio sounds; each platform expects a set level.
Inside the OS
Where the audio engines work
Voiceovers, podcasts, music, phone greetings, sound effects and audiograms.
Music ducked under the voice, the master set to −16 LUFS, the voice cleaned up.
A voiceover pass for every scene you make.
Pair a voice to a face, so your host sounds the same every time.
The voice that answers your phone, mastered for the line.
1 podcast becomes many audiograms, sized for each channel.
Put 2 voices on 1 script and hear both plans before you pay.
Your signature sound, saved once and used everywhere.
Free engines first, always. Metered engines show cost and ask first. Quoted on the sum, charged on what returned, released on cached and skipped work. Engines without a commercial license are blocked, including a popular non-commercial music model. Suno vocal songs carry a rights flag over music-industry lawsuits (RIAA); you decide before one ships.
Questions owners ask about audio
Can I use my own voice?
Yes. Record a signed release first, then the OS clones it. The release is enforced in code; withdraw consent and the voice stops being used.
Is the music safe to use in my ads?
Free instrumentals come from commercially cleared engines; sound effects from a CC0 or CC-BY library, credit tracked. Vocal songs carry a rights flag so you decide first.
What does audio cost me?
Free-lane work runs at $0 per job. Metered engines like ElevenLabs show cost and ask first. Nothing paid is a silent default.
Do these voices answer my phone?
The receptionist voice comes from this lane (Piper and NeuTTS, or ElevenLabs if chosen). Built in, live once your phone line is connected. See phone engines.
How many languages?
The Scoring Stage works in 4 languages today, with speeds from 0.75× to 1.5×.
Keep going
Related pages
Try the OS
Our free trial is starting soon.
The same tools we use for our done-for-you clients. Our 7-day free trial is starting soon. Until then, book a call and a real person walks you through the OS. $997 a month, first month $497. No contracts.

YOUR TEAM runs it with the OS, or OUR TEAM runs it for you with Done-For-You. Not sure? Which one fits me?





