Frequently Asked Questions
Everything you need to know about our AI voiceover, dubbing, music production, and speech data services.
About Onyx Studios
What is Onyx Studios?
Onyx Studios (onyxstudios.ai) is a professional AI voice and music production studio founded in 2008, headquartered in Taiwan. We offer AI voiceover, multilingual AI dubbing, AI-assisted music production, and enterprise speech data collection. All AI output is reviewed and refined by experienced human voice directors before delivery.
How do I get started? Can I pay online, or should I enquire first?
Both paths work. If you already know what you need, online checkout is the fastest route — submit your script, select a plan, complete payment, and production is scheduled immediately, typically delivered within 24 hours. If you would like to discuss your requirements first, explore custom options, or have a large-scale or enterprise project, email us at support@onyxstudios.ai or use the contact form. We respond within one business day and can handle everything offline from there.
What makes Onyx Studios different from other AI voice platforms?
Onyx Studios is not a self-serve TTS tool. Every project is handled end-to-end by a production team: AI generates the first pass, then human voice directors refine tone, pacing, pronunciation, and emotion for delivery-ready results. Our voice models are built from recordings by contracted, consenting voice actors — not scraped or anonymous data.
AI Voiceover
How much does AI voiceover cost?
AI voiceover plans start from $39 (0–60s AI Instant Voice package). Pricing depends on word count, language, delivery speed, and whether human director polish is included. See full tier breakdowns at onyxstudios.ai/pricing.
How long does delivery take?
Standard delivery is within 24 hours for most AI voiceover orders. Rush (same-day) delivery is available on select plans. Large-scale projects — multilingual dubbing or speech data collection — have custom timelines discussed at the briefing stage.
How many revisions do I get? What if I'm not satisfied?
Revision allowances vary by plan — see each tier's details on the pricing page. Revisions cover re-generation with adjusted direction notes (tone, pacing, pronunciation). If a delivery has objective errors — mispronounced words, missing lines, or technical quality issues — we redo it at no charge, regardless of revision count. Completed deliveries are non-refundable; if your project has specific requirements, state them clearly in your brief before production begins so both sides share the same expectations.
Is AI-generated voiceover as good as a human recording?
For narration, e-learning, ads, and branded content, our human-directed AI voiceover is virtually indistinguishable from a studio recording in most production contexts. We use proprietary voice models trained on professional studio recordings, combined with human director review. For highly emotional or character-driven performance, a 100% human studio recording upgrade is also available.
What is human-directed AI voice production?
Human-directed AI voice production is our core methodology: AI synthesizes the audio draft, then a trained voice director listens through, adjusts pacing, catches mispronunciations, re-generates problem segments, and signs off on the final file. It combines the speed and cost efficiency of AI with the quality assurance of a seasoned studio professional.
What languages does Onyx Studios support?
Onyx Studios covers 16+ major commercial languages: English (US / UK), Mandarin (Traditional / Simplified), Cantonese, Japanese, Korean, Spanish, French, German, Portuguese, Italian, Russian, Arabic, Hindi, Indonesian, Vietnamese, and Thai. Rare languages (Dari, Swahili, Burmese, etc.) are available on request.
Can I request a specific accent within a language?
Yes. When ordering or submitting a brief, you can specify the accent or regional variant you need — for example, US English vs. UK English, Taiwanese Mandarin vs. Mainland Mandarin, or Brazilian Portuguese vs. European Portuguese. Our roster covers the most common regional variants. If a specific accent is critical to your project, mention it early so we can confirm availability before production begins.
Can you clone a specific voice or build a custom voice model?
Yes, with conditions. Onyx Studios can build a custom AI voice model from recordings of a consenting, contracted voice actor — whether that is your brand spokesperson, an internal narrator, or a talent sourced through our network. The voice actor must provide explicit written consent and sign a licensing agreement before any model training begins. We do not clone voices without documented consent. Custom voice development is quoted separately; contact us with your requirements for a feasibility assessment.
What commercial rights do I receive with my order?
All standard orders include full commercial rights for the licensed use case — ads, branded video, e-learning, internal communications, and digital distribution — plus a verifiable License Certificate. Broadcast rights, perpetual global rights, and resale rights are available on higher-tier plans or via custom licensing.
What file formats do you deliver?
Voiceover deliverables are provided as broadcast-quality WAV + MP3 by default. Speech data and live strings recording projects default to 48 kHz / 24-bit WAV. Dubbing projects include dubbed audio tracks and, where applicable, subtitle files (SRT / VTT). If you have specific format requirements, include them in your brief.
AI Dubbing
How does Onyx Studios preserve the original actor's voice in dubbing?
Our AI dubbing technology clones the tonal fingerprint, pitch contour, and vocal texture of the original speaker, then synthesizes that voice speaking the translated script in the target language. The audience hears the original actor's voice identity — not a generic TTS voice — even in a completely different language version.
What is lip-sync dubbing, and will my video length change?
Lip-sync dubbing aligns the timing of dubbed audio to the on-screen mouth movements of the original actor so speech appears natural when viewed. Onyx Studios uses AI-assisted lip-sync to match phoneme duration and phrasing to the original video without re-editing the picture. One important note: because different languages vary in syllable structure and sentence length — for example, English phrases are often longer than their Mandarin equivalents — the total duration of a dubbed video may differ slightly from the original. A typical drift is a few seconds to around ten seconds per four minutes of content (e.g., a 4:00 original may become a 4:10 dubbed version). This is a normal characteristic of cross-language dubbing, not a quality issue, and is standard across the industry. If you have strict runtime requirements, please let us know at the briefing stage and we will assess feasibility.
Does dubbing include translation, or do I need to provide the script?
By default, clients provide the translated, approved script and we produce from it. If you need translation included, it can be arranged as a paid add-on — include the source language, target language, and content type in your brief and we will provide a translation quote alongside the production quote.
Music Production
What is AI-assisted music production at Onyx Studios?
Onyx Studios offers hybrid AI music production: AI tools generate arrangement sketches and direction demos, which senior producers then develop into full production — arrangement, orchestration, mixing, and mastering. The result is original, production-ready music at a fraction of the time and cost of traditional studio composition.
Who owns the music I commission?
It depends on the plan. AI Curator and Pro Arrangement plans include a perpetual commercial license upon full payment — covering streaming, advertising, sync, and broadcast — but the underlying composition copyright remains with Onyx Studios, and your commissioned piece is never re-sold to other clients. The Masterpiece tier is a full copyright buyout (work-for-hire): complete ownership transfers to you. Stems and source files are available from Pro Arrangement upward. See each plan's details for the exact licensing scope.
Speech Data
What is AI speech data collection? Do you offer annotation?
AI speech data collection covers recording, segmenting, and transcribing large volumes of human voice audio for TTS and ASR model training. Onyx Studios' core deliverable includes talent recruitment and session direction, audio QC, spec-compliant segmentation, transcription, and basic metadata tagging — datasets delivered ready for training pipelines. Annotation is available as an add-on, subject to language and complexity. Standard tasks such as timestamp alignment and speaker identification are generally feasible; more complex annotation types — emotion labeling, multi-dimensional phonetic classification, and similar — are evaluated on a case-by-case basis. If annotation is part of your requirement, include the spec in your brief and we will assess feasibility and provide a separate quote.
Enterprise & Legal
Do you work with enterprise clients?
Yes. Onyx Studios works with advertising agencies, streaming platforms, e-learning companies, game publishers, and AI companies building proprietary voice models. Enterprise clients can negotiate custom delivery guarantee terms (SLA), NDAs, and dedicated production capacity. Contact support@onyxstudios.ai for enterprise inquiries.
Is my content kept confidential? Do you sign NDAs?
Yes. All client scripts, project briefs, and deliverables are treated as confidential and are not used for any purpose beyond the contracted scope. If you require a formal NDA before sharing materials, we are happy to sign one — contact us before submitting your brief and we will arrange it promptly.
Are the AI voices at Onyx Studios legally licensed?
Yes. Every AI voice at Onyx Studios is built from recordings by consenting, contracted voice actors who have been compensated for their contribution. We do not use scraped, unauthorized, or anonymous voice data. Each voice model is covered by a proprietary licensing agreement, and every client order includes a verifiable License Certificate.
Still have questions?
Our team is ready to help. You can also submit a project brief directly and we'll get back to you within one business day.
Contact Us