What Is TTS API Voice Over For Video Ads?

By Rn98 editorial team · Updated 2026-11-19

This page explains how SpotVoce’s TTS API replaces manual voice-over hand-offs for advertising agencies and radio production teams. We provide a standardized endpoint that integrates directly into your existing creative workflows, allowing you to generate programmable voice-over for video and audio spots without external dependencies.

Readers will learn how to request speech via simple JSON payloads, manage voice presets, and handle audio delivery. We cover technical requirements, integration patterns with video editors, and the specific constraints of our engine so you can accurately estimate production speed and costs for your next campaign.

What Is TTS API Voice Over For Video Ads?

Frequently asked questions

What does TTS stand for?

TTS stands for Text-to-Speech. It is a technology that converts written text into spoken audio files. Our API accepts text input and returns synthesized audio, enabling automated voice-over generation for video ads without manual recording sessions.

How does TTS differ from voice actors?

Unlike human voice actors, TTS generates audio programmatically via an API. It offers consistent quality, instant delivery, and no scheduling constraints. However, it lacks the nuanced emotional range of human performers and produces audio with an inaudible synthetic watermark.

Can I use TTS for commercial ads?

Yes. The generated audio files are suitable for commercial use in video ads and radio spots. Input text is used only to produce the audio and is not used for training. You receive the audio file directly via the API response.

Is the voice synthetic or real?

The voice is synthetic. It is based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobook readings. Every generated file carries an inaudible watermark that marks it as synthetic speech, distinguishing it from live human recordings.

How long does it take to generate audio?

Generation takes roughly real time, about 25 characters of text per second of audio. Short phrases may take a few seconds. Requests are processed sequentially, so under load, a request may wait in a queue before synthesis begins.

Do I need technical skills to use it?

Basic technical skills are helpful but not mandatory. The API follows the OpenAI request shape, allowing use of official OpenAI SDKs by changing the base URL. You can also use ready curl examples provided on the /get-api-key/ page for quick integration.

Can I change the voice after generation?

No. Once the audio file is generated and returned, the voice is fixed. You must specify the desired voice (e.g., aria, nova, felix) in the request parameters before generation. Regenerating with a different voice requires a new API call and incurs additional cost.

Is the audio copyright-free?

The voices are based on LibriTTS-R (CC BY 4.0) and LibriVox public-domain readings. However, the generated audio files carry an inaudible watermark marking them as synthetic speech.

Can I use the API for radio spots?

Yes. The API supports formats like mp3, wav, opus, flac, and pcm, suitable for radio production. You can select voices like atlas (neutral, news) or felix (energetic, promo) to match the tone of your radio spot efficiently via the API.

What is the difference between tts-1 and tts-1-hd?

Both models use the same engine and price. The only difference is the model identifier in the request. Use 'tts-1' for standard generation or 'tts-1-hd' for high-definition audio. Both incur the same cost of $10 per 1 million characters.

Does the API support emotion?

No. The API does not support emotion tags or custom voice design. Voices have preset characteristics (e.g., warm, bright, calm), but you cannot dynamically adjust emotional tone during generation. You select a voice preset that best matches the desired tone.

Can I pause and resume generation?

No. The API does not support streaming or WebSocket connections. Each request is processed independently. If you need to pause, you must cancel the request or wait for it to complete. There is no partial audio stream to resume from.

Is there a limit on daily usage?

There is no daily limit, but there is a rate limit of 60 requests per minute per key. Longer texts must be split into parts of up to 4096 characters. Failed requests are free, but successful ones count against your character balance.

How do I integrate this with Premiere Pro?

The API does not offer a native Premiere Pro plugin. You can generate audio files via the API using curl or a script, then import the resulting mp3 or wav files into Premiere Pro. Use the /docs/ page for request details.

Definition of TTS API

Sound wave illustration for: Definition of TTS API

Text-to-Speech (TTS) technology converts written text into synthesized audio files via a cloud API. Instead of booking voice actors, developers send JSON payloads to receive MP3, WAV, or FLAC files instantly. This eliminates manual hand-offs between copywriters and production teams, allowing your creative workflow to remain entirely digital and automated.

The integration model mirrors the OpenAI speech endpoint structure, ensuring compatibility with standard HTTP clients and SDKs. By pointing your base URL to https://api.rn98.info/v1, you utilize the OpenAI SDKs with minimal code changes. This standardized approach reduces integration friction for engineering teams already familiar with modern API architectures.

Text-to-Speech Technology

Our engine generates high-fidelity audio from text inputs up to 4096 characters per request. The synthesis process is fully programmatic, allowing you to trigger voice generation directly from your build pipelines or content management systems without human intervention.

API Integration Model

Integration requires only a POST request to the /v1/audio/speech endpoint. Authentication is handled via a single Bearer token, simplifying credential management in your CI/CD environments. The response returns the audio file directly, ready for immediate assembly in your editing timeline.

Programmatic Control

You control every aspect of generation through JSON parameters: select from 8 distinct voice presets, adjust speed between 0.25x and 4.0x, and specify the output format. This granular control ensures consistent audio characteristics across thousands of generated spots without manual adjustment.

Application in Advertising

Sound wave illustration for: Application in Advertising

TTS API voice over for video ads uses programmatic text-to-speech generation to produce audio tracks directly from scripts. By sending JSON payloads to the API, advertising agencies can automate voice-over creation without manual hand-offs between copywriters and voice talent.

This method integrates seamlessly into existing creative workflows, allowing teams to generate consistent, standardized audio assets programmatically. It eliminates scheduling bottlenecks and enables rapid iteration of ad copy with precise voice selection and speed control.

Voice Over Generation

The API accepts up to 4096 characters per request, converting text into high-quality audio files in formats like MP3 or WAV. With eight preset voices ranging from warm narrators to energetic promo styles, agencies can match specific brand tones without hiring multiple voice actors.

Each generated file carries an inaudible synthetic watermark, ensuring clear identification of AI-generated content while maintaining professional audio standards suitable for broadcast and digital platforms.

  • Generate audio from plain text via JSON POST requests
  • Choose from 8 distinct voice presets for tone matching
  • Support for 23 languages with automatic language detection

Video Content Creation

TTS API enables the creation of voice-overs for explainer videos, social media ads, and e-commerce product spots. Teams can update ad copy instantly and regenerate audio without re-recording, ensuring content remains fresh and relevant.

The API’s speed of roughly 25 characters per second allows for near real-time generation of longer scripts, facilitating quick assembly of video assets directly from script repositories or CMS platforms.

  • Instant audio generation for social media and broadcast ads
  • Easy integration with video editing timelines
  • Consistent voice across multiple video campaigns

Rapid Prototyping

Agencies can quickly prototype ad variations by swapping scripts and voices programmatically. This allows for A/B testing of different voice tones and messages without the cost and delay of booking studio time.

Failed synthesis requests are free, reducing risk during iterative development. Teams can experiment with multiple voice options and scripts, refining the final audio asset before committing to full production.

  • Test multiple voice options and scripts instantly
  • No cost for failed synthesis requests
  • Reduce time-to-market for ad campaigns

Advantages Over Traditional Methods

Sound wave illustration for: Advantages Over Traditional Methods

Traditional voice-over production involves casting, studio booking, and post-production editing, often taking days. Our TTS API generates audio programmatically, reducing production time from days to seconds per request. This allows agencies to iterate creative concepts rapidly without logistical delays.

Cost efficiency stems from a simple per-character model at $0.01 per 1,000 characters. There are no hourly studio rates, session fees, or minimum spend requirements. Failed requests are free, ensuring you only pay for usable assets.

Speed of Production

Generate voice-overs in seconds rather than days. The API processes requests sequentially, delivering audio files quickly even for long scripts. This speed enables rapid prototyping and A/B testing of ad variations without rescheduling talent.

Traditional workflows require coordinating schedules and editing studios. Our programmatic approach integrates directly into your creative pipeline, eliminating manual hand-offs and allowing immediate generation of multiple script versions.

  • Generate audio in seconds, not days
  • No studio booking or scheduling delays
  • Immediate iteration for A/B testing
  • Sequential processing of multiple scripts

Cost Efficiency

Pay only for characters used at $0.01 per 1,000 characters. No monthly retainers or minimum commitments. Failed requests are refunded in full, ensuring zero waste on unusable audio. This model scales directly with your volume.

Eliminate fixed costs like studio rentals and session fees. With no minimum spend, small campaigns remain affordable. The absence of middlemen reduces overhead, making high-quality voice-over accessible for any budget size.

  • $0.01 per 1,000 characters
  • No monthly retainers or minimums
  • Zero cost for failed requests
  • No studio or session fees

Scalability

Handle volume growth effortlessly. Generate hundreds of ad variations without hiring additional voice talent or producers. The API supports up to 60 requests per minute, ensuring consistent throughput for large campaigns.

Scale production up or down based on demand. With no seat limits or capacity constraints, your creative team can produce as many assets as needed. Integration into existing workflows ensures seamless expansion without operational bottlenecks.

  • No seat or capacity limits
  • Supports high-volume generation
  • Consistent throughput under load
  • Seamless workflow expansion

Technical Requirements

Sound wave illustration for: Technical Requirements

The API requires a stable internet connection for all synthesis requests. Input text is sent to the server, and the resulting audio file is returned in the response. No local processing occurs, so latency depends on network stability and server queue status.

Basic developer knowledge is needed to handle HTTP POST requests and manage JSON payloads. You must understand how to pass voice IDs, model types, and text inputs. Standard OpenAI SDKs work by adjusting the base URL, simplifying integration for existing engineering teams.

Internet Dependency

All synthesis happens on the server. A reliable connection is required to send the text payload and receive the audio file (mp3, wav, opus, flac, or pcm). Latency varies with queue load, so plan for slight delays during peak usage.

No local dependencies or heavy computing resources are needed on your end. The processing power is entirely cloud-based, keeping your local environment lightweight.

  • Stable internet connection required.
  • Server-side processing only.
  • No local software dependencies.

Developer Knowledge

Integration requires basic API literacy: constructing HTTP POST requests with JSON bodies. You need to manage the base_url, model, voice, and input fields. Standard OpenAI SDKs work seamlessly by pointing to https://api.rn98.info/v1.

Understanding character limits (4096 chars per request) is crucial for longer scripts. You must implement client-side splitting if your text exceeds this limit, then join the resulting audio files.

  • HTTP POST requests with JSON.
  • OpenAI SDK compatibility.
  • Character limit management (4096 chars).

API Key Management

Authentication uses a single Bearer token sent in the Authorization header. One active key per account ensures clear usage tracking. If a key is compromised, you can issue a new one instantly, and the old one stops working immediately.

Keys are generated via the /get-api-key/ dashboard. Monitor usage and balance directly in the account panel to avoid service interruption due to insufficient credits.

  • Single Bearer token per account.
  • Instant key rotation on revocation.
  • Dashboard-based usage monitoring.

Voice Selection Process

Sound wave illustration for: Voice Selection Process

Selecting a voice is the first step in defining the auditory identity of your spot. SpotVoce provides eight preset voices, each designed for specific tonal roles, allowing you to choose between warm narrators, energetic promo speakers, or neutral news readers without manual casting.

The selection process is programmatic. You pass the voice identifier in the API request, ensuring consistent audio output across all generated assets. This eliminates the variability of human talent and guarantees that every ad maintains the exact vocal character you defined.

Preset Voice Options

We offer eight distinct voices: Aria, Nova, Luna, and Sage for female tones; Felix, Milo, Atlas, and Orion for male tones. Each voice is derived from licensed recordings, providing a standardized, high-quality baseline for your production needs.

These presets cover a range of emotional registers, from the soft, storytelling quality of Luna to the deep, documentary tone of Orion. You can also adjust the speed between 0.25x and 4.0x to fit your video’s pacing requirements.

  • Aria: Warm, clear narrator
  • Felix: Energetic, promo
  • Atlas: Neutral, news
  • Orion: Deep, documentary

Matching Brand Tone

Choose a voice that aligns with your brand’s existing audio identity. For a tech product, Atlas provides a neutral, trustworthy tone. For a lifestyle brand, Nova’s bright, upbeat character may be more appropriate.

If your script requires a conversational feel, Milo offers a friendly, approachable delivery. This consistency helps viewers instantly recognize your brand across different campaigns and media channels.

Audience Targeting

Voice selection can influence audience perception. A younger demographic might respond better to the energetic tone of Felix, while an older audience might prefer the calm, instructional clarity of Sage.

Test different voices with your target segment to see which drives better engagement. Since generation is rapid, you can quickly prototype multiple variations and select the most effective one for your final ad.

Audio Output Specifications

Sound wave illustration for: Audio Output Specifications

SpotVoce returns synthesized audio via a standard HTTP response, delivering the final asset directly for immediate use in your ad pipeline. The API supports five distinct output formats—mp3, wav, opus, flac, and pcm—allowing you to select the optimal balance between file size and fidelity for your specific delivery channel or editing software.

File Formats

Choose mp3 for compact web delivery, wav for uncompressed high-quality editing, or opus/flac for efficient storage. PCM provides raw data for custom processing. All formats are generated instantly upon successful synthesis completion.

  • mp3: Ideal for web and social media delivery.
  • wav: Uncompressed, best for post-production editing.
  • opus/flac: High efficiency with lossless or near-lossless quality.
  • pcm: Raw audio data for custom pipelines.

Audio Quality

Output quality is determined by the model: 'tts-1' or 'tts-1-hd' use the same engine and price. Each file includes an inaudible synthetic watermark for attribution. Quality remains consistent across all 23 supported languages without degradation.

Duration Estimation

Generation speed is approximately real-time for long texts (25 characters per second). Short phrases may take a few seconds due to queue processing. Total duration matches the spoken length of your input text exactly, up to 4096 characters per request.

Integration with Video Editors

Sound wave illustration for: Integration with Video Editors

Integrating SpotVoce into your video production pipeline eliminates manual hand-offs by automating voice-over generation. Instead of waiting for studio schedules, your creative team triggers audio synthesis directly from existing scripts, ensuring that visual edits and voice tracks stay synchronized without version control issues.

Workflow Automation

Connect the TTS API to your editing timeline via simple HTTP triggers. When a scene is finalized, the system generates the audio file and delivers it to your asset server, ready for immediate import into Premiere Pro or DaVinci Resolve.

  • Trigger synthesis via scene completion events
  • Auto-download audio to shared project folders
  • Sync voice timing with visual cuts programmatically

API Calls from Scripts

Use Python or JavaScript wrappers to send text to https://api.rn98.info/v1/audio/speech directly from your editing scripts. Pass the voice and speed parameters to match the director’s creative notes, receiving an MP3 or WAV file that matches the exact length of the cut.

  • Python/JS scripts send JSON payloads
  • Receive audio files via HTTP response
  • Adjust speed (0.25–4.0) programmatically

Asset Management

Store generated audio alongside video assets in your DAM. Since each request produces a unique file, you can tag audio with the corresponding scene ID, voice preset, and language code, making it easy to audit which voice was used for each ad variant.

  • Tag audio with scene ID and voice
  • Store in existing DAM or NAS
  • Link audio to specific ad versions
Integration with Video Editors

Language and Localization

Sound wave illustration for: Language and Localization

SpotVoce supports 23 languages, from English and Spanish to Arabic and Japanese. Any voice preset can deliver any supported language, allowing a single API key to handle global campaigns without managing multiple voice assets.

Multi-Language Support

The API detects language from the input script by default, but you can explicitly set the ISO code to ensure accuracy. This is critical for Latin-script texts that might otherwise default to English. All 23 supported languages, including Swahili and Turkish, are fully compatible with all eight voice presets.

Dubbing Workflows

Generate distinct audio tracks for each target market by sending localized text to the same endpoint. Since the API returns standard audio files (mp3, wav, opus, flac), you can easily sync these tracks with your video editors. This eliminates the need for separate recording sessions or studio bookings for each language variant.

Accent Variations

Voices are based on licensed recordings and do not offer adjustable accent sliders or phoneme-level control. Each voice has a consistent tonal character regardless of the language spoken. For example, Aria delivers a warm narrator tone in both English and French. If a specific accent variation is required, you must select the voice that best matches that tonal profile.

  • Supported: Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish.
  • Not supported: Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, Bengali.

Quality Assurance

Sound wave illustration for: Quality Assurance

Quality assurance for synthetic voice-over relies on consistent technical output and human validation. Each generated audio file includes an inaudible watermark, marking it as synthetic speech. This allows for easy identification during post-production without affecting the audio quality or listener experience.

Synthetic Watermark

The watermark is embedded directly into the audio data. It serves as a clear marker for broadcasters and platforms that require disclosure of AI-generated content. This ensures transparency without adding any audible artifacts to the final spot.

Consistency Checks

We verify that the speed and voice parameters match the request exactly. Since requests are processed sequentially, we monitor for queue delays that might affect timing. This ensures the audio aligns perfectly with your video timeline.

Character counts are validated against the 4,096-character limit per request. Failed requests are refunded, ensuring you only pay for delivered audio. This reduces risk in high-volume ad production cycles.

Human Review Process

Final approval remains with your creative team. The API provides raw audio files that you can review in your standard DAW or video editor. This allows for quick adjustments to speed or voice selection before finalizing the spot.

Since there is no automatic emotion tagging, humans review the tone. This is critical for advertising, where nuance matters. You control the final cut, ensuring the synthetic voice meets your brand’s specific standards.

Use Case Examples

Sound wave illustration for: Use Case Examples

SpotVoce enables agencies to generate voice-overs programmatically, removing manual hand-offs between copywriters and voice actors. By sending text via our API, teams receive synthesized audio files ready for immediate integration into video editing timelines.

Social Media Ads

Generate rapid variations of ad copy for A/B testing on platforms like TikTok or Instagram. Use distinct voices such as 'Nova' for upbeat promo spots or 'Atlas' for neutral news-style updates, ensuring consistent branding across multiple creatives without scheduling conflicts.

  • Rapid A/B testing of voice tones
  • Consistent brand voice across channels

Explainer Videos

Convert script drafts directly into audio using 'Sage' for instructional clarity or 'Luna' for storytelling. This allows developers to update video narration instantly when copy changes, eliminating re-recording costs and keeping production cycles tight.

  • Instant script-to-audio updates
  • Clear instructional narration

E-Commerce Product Videos

Automate voice-overs for dynamic product listings. Integrate API calls into your content management system to generate unique audio for each SKU, ensuring accurate descriptions without manual recording for every new item.

  • Automated SKU narration
  • Scalable content generation

Limitations to Consider

Sound wave illustration for: Limitations to Consider

Understanding the constraints of our TTS API ensures accurate planning for your ad production pipelines. While our engine handles standard commercial copy efficiently, specific technical boundaries apply to character count, processing latency, and voice customization options.

Character Limits

Each API request accepts a maximum of 4096 characters per input. Longer scripts must be split into segments and joined programmatically after synthesis.

Processing Time

Synthesis typically generates audio at roughly real-time speed (approx. 25 characters per second). Under high load, requests queue sequentially, potentially adding latency beyond the generation time itself.

No Custom Cloning

We provide eight fixed voice presets (e.g., Aria, Felix, Atlas) based on licensed recordings. We do not offer custom voice cloning, emotion tags, or SSML support, ensuring consistent, standardized output across all campaigns.

Future Scalability

Sound wave illustration for: Future Scalability

SpotVoce’s architecture supports your account’s growth by scaling with your campaign volume. Because requests are processed sequentially, higher volumes may introduce queue delays during peak loads, ensuring consistent output quality without overloading the engine. This linear scaling model means you can predict processing times based on your batch size, allowing your production team to plan render cycles accurately.

Volume Growth

Your usage scales directly with your ad output needs. As you produce more spots, the system handles the load sequentially. There are no complex concurrency limits to manage, just a straightforward queue system that maintains stability even during high-demand periods.

  • Sequential processing ensures stability under load
  • No complex concurrency management required
  • Predictable scaling based on batch size

API Rate Limits

The API allows 60 requests per minute per key. This limit is sufficient for most agency workflows, especially when splitting longer scripts into 4,096-character segments. If you exceed this, requests wait in the queue rather than failing, preventing data loss or corruption during high-volume pushes.

  • 60 requests per minute per key
  • Queuing prevents request loss
  • Easy to manage with segmented scripts

Cost Projection

Pricing is strictly per character at $10 per 1 million characters. This transparency allows you to forecast costs precisely. For example, a 60-second ad script (approx. 150 characters) costs $0.0015. Your balance never expires, and unused credit remains available for future campaigns, making long-term budgeting straightforward.

  • $10 per 1 million characters
  • Predictable per-script costs
  • Balance never expires

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. Her gentle tone reduces listener fatigue during longer audio spots.

Avatar of the voice aria
ariafemale
Avatar of the voice nova
novafemale
Avatar of the voice luna
lunafemale
Avatar of the voice sage
sagefemale
Avatar of the voice felix
felixmale
Avatar of the voice milo
milomale
Avatar of the voice atlas
atlasmale
Avatar of the voice orion
orionmale

23 languages

Any voice speaks any of these languages.

Sources and further reading

  1. Wikipedia
  2. openai-python on GitHub
  3. Creative Commons Attribution 4.0
  4. openai-node on GitHub
  5. LibriTTS-R corpus (OpenSLR 141)
  6. ISO 639 language codes
  7. IETF RFC 7845: Ogg Opus
  8. MDN: Web Speech API
  9. W3C WAI: making audio and video accessible
  10. Tether (USDT)
  11. FFmpeg documentation
  12. curl documentation

More on this site