How to Generate Radio Ads with TTS API? | SpotWave

By SpotWave editorial team · Updated 2026-11-18

This guide details how ad agencies and radio producers can integrate the SpotWave TTS API to generate broadcast-ready voice-overs at scale. We explain the technical workflow from script segmentation to audio delivery, ensuring your spots meet character limits and timing requirements.

You will learn to select appropriate voices, handle bulk synthesis efficiently, and manage licensing for commercial use. The page covers integration specifics, including rate limits and format selection, providing concrete data for production teams evaluating API solutions for their next radio campaign.

How to Generate Radio Ads with TTS API? | SpotWave

Frequently asked questions

What is the best voice for a radio spot?

Q: Which voice is best for a radio spot? A: Felix is designed as an energetic, promo voice ideal for radio spots. Atlas offers a neutral, news-style tone suitable for clear announcements. Milo provides a friendly, conversational sound for relatable messaging. Choose based on your brand’s specific tone, as all voices support 23 languages and can be adjusted for speed between 0.25 and 4.0.

How long does it take to generate a 30-second ad?

Generation takes roughly real-time, averaging 25 characters per second. A standard 30-second radio ad contains approximately 60–75 words. Therefore, synthesis typically completes within 30–45 seconds. Requests are processed sequentially, so under heavy load, your request may wait in a queue before processing begins.

Can I edit the audio after generation?

The API returns a complete audio file (mp3, wav, opus, flac, or pcm). You can edit this file using standard audio software. The API does not support SSML or phoneme-level editing, so you must adjust timing and tone externally. Generated audio is not stored by the API.

Do I need a microphone for API voice overs?

No. The API generates synthetic speech using licensed recordings from LibriTTS-R and LibriVox. You do not need recording equipment or voice talent. Each file includes an inaudible watermark identifying it as synthetic speech, ensuring transparency for your broadcast or digital ad placements.

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

Both models use the same engine and price ($0.01 per 1,000 characters). The only difference is the output quality setting. Select tts-1-hd for higher fidelity audio. Use tts-1 for standard quality. Both support the same 8 voices and 23 languages with identical speed and format options.

Can I use the audio in YouTube ads?

Yes. The voices are based on licensed recordings (LibriTTS-R CC BY 4.0) and public-domain readings. You can use the generated audio in YouTube ads, radio spots, or other commercial content. The API handles the synthesis; you retain the rights to the generated audio files for your campaigns.

How do I handle punctuation in scripts?

Punctuation counts as characters in your usage billing. Use standard commas, periods, and ellipses to control pacing and pauses. The engine interprets these naturally. There is no SSML support, so complex pause durations must be managed by adjusting text length or using the speed parameter.

What is the character limit per request?

Each request supports up to 4,096 characters of input text. For longer radio scripts, split your text into parts of 4,096 characters or less. Generate each part separately and join the resulting audio files in your editing software. Spaces and punctuation are included in this count.

Can I change the speed of the voice?

Yes. The API allows speed adjustment between 0.25 and 4.0. You can slow down the narration for clarity or speed it up for energetic promos. This applies to all 8 voices. Adjusting speed affects the total character-per-second ratio, helping you fit specific time slots.

Do the voices sound robotic?

The voices are based on licensed human recordings from LibriTTS-R and LibriVox. They offer natural intonation, warmth, and clarity. While synthetic, they are designed for professional use in ads and radio. Each file carries an inaudible watermark to mark it as synthetic speech, ensuring transparency.

Is there a limit on the number of words?

The limit is 4,096 characters per request, not words. Since spaces count as characters, this allows roughly 800–900 words per request. For longer scripts, split the text into multiple requests. Each request is billed separately based on the character count used.

Can I use multiple voices in one ad?

The API generates one voice per request. To use multiple voices, generate each segment separately with the desired voice parameter, then combine the audio files in your editing software. You can switch voices between sentences or paragraphs by making separate API calls for each segment.

How do I test the API before paying?

Sign in with Google to receive a free trial with $0.50 credit (approx. 50,000 characters), valid for 7 days. Use this to generate test audio. Failed requests are free. If you use email/password, you start without trial credit and must top up via USDT (TRC-20) to begin.

What formats are best for radio?

MP3 and WAV are standard for radio broadcasting. MP3 offers smaller file sizes for easy transfer. WAV provides uncompressed quality. You can also choose OPUS, FLAC, or PCM. Select the format that matches your radio station’s submission requirements or your editing workflow’s preferences.

Why use API for radio spots

Sound wave illustration for: Why use API for radio spots

Using a TTS API for radio spots eliminates the overhead of studio booking and talent fees. You pay only for what you generate at $0.01 per 1,000 characters, making it highly cost-efficient for agencies producing high volumes of ad variants.

The API delivers audio files directly via JSON responses, removing the wait time for studio booking. This direct integration allows for rapid iteration of scripts and tones, ensuring your radio spots are ready for broadcast quickly.

Cost efficiency

Traditional voice-over production involves fixed costs for studio time and talent rates, regardless of the final audio length. With SpotWave, you pay strictly for character count. At $0.01 per 1,000 characters, generating a 30-second spot costs pennies, allowing you to test multiple voice angles without budget fatigue.

  • Pay-per-use model: $0.01 per 1,000 characters
  • No minimum spend or monthly retainers
  • Failed requests are free
  • No hidden studio or engineer fees

Speed of delivery

Human voiceovers require scheduling, recording, and editing, often taking days. The API generates audio in seconds for short scripts, but longer texts process at roughly real-time speed. While longer texts process at roughly real-time speed, short ad scripts are synthesized almost instantly, enabling same-day delivery for time-sensitive campaigns.

  • Instant generation for short scripts
  • No scheduling delays with talent
  • Direct JSON audio response
  • Rapid iteration of multiple versions

Scalability

Agencies often need to produce dozens of localized or A/B tested versions of a single spot. The API handles bulk synthesis seamlessly. You can split long scripts into 4,096-character segments and join the audio, ensuring consistent quality whether you produce one spot or one hundred.

  • Supports 8 distinct voice presets
  • Handles 23 languages natively
  • Batch processing for multiple variants
  • Consistent audio quality across volume

Choosing the right voice

Sound wave illustration for: Choosing the right voice

Selecting the appropriate voice preset is critical for radio spot conversion. SpotWave provides eight distinct presets, each designed for specific tonal requirements in advertising.

Warm narrator style

Aria offers a warm, clear narration ideal for storytelling segments. Luna provides a soft, storytelling tone suitable for narrative-driven spots.

  • Aria: warm, clear narrator

Energetic promo style

Felix delivers an energetic, promo-ready voice perfect for high-impact advertisements. Nova provides a bright, upbeat tone suitable for fast-paced promotional content.

  • Felix: energetic, promo

Neutral news style

Atlas ensures a neutral, news-style delivery for clear, factual announcements. Sage offers a calm, instructional tone suitable for informational spots.

  • Atlas: neutral, news

Script preparation

Sound wave illustration for: Script preparation

Radio spots demand precision. Every character counts toward your budget, and improper formatting can ruin the audio flow. Proper preparation ensures your voice-over matches the intended pacing and tone without wasting credits on retries.

Start by calculating the exact character count, including spaces and punctuation. This determines the cost ($0.01 per 1,000 characters) and helps you stay within the 4,096-character limit per request.

Character count calculation

Count every character, including spaces. This ensures accurate cost estimation and prevents unexpected charges. Use a simple text editor or script to verify the length before sending it to the API.

  • Include spaces and punctuation in the count
  • Verify length is under 4,096 characters
  • Use a text editor for accurate counting

Handling punctuation

Punctuation marks affect the voice’s pacing. Commas, periods, and ellipses create natural pauses. Ensure your script uses clear punctuation to guide the TTS engine, avoiding awkward phrasing or rushed delivery.

For complex scripts, test small segments first. Adjust punctuation to achieve the desired tone and timing. This iterative approach helps fine-tune the audio output before committing to the full spot.

  • Use commas for short pauses
  • Periods for complete stops
  • Ellipses for trailing thoughts

Segmenting long scripts

If your script exceeds 4,096 characters, split it into smaller segments. Generate each segment separately and join the audio files afterward. This ensures consistent quality and avoids truncation errors.

Plan your segments carefully. Each part should be a coherent phrase or sentence to maintain the narrative flow. Test the join process to ensure seamless audio transitions.

  • Split scripts into 4,096-character chunks
  • Ensure each segment is coherent
  • Test audio joins for smooth transitions

Integration process

Sound wave illustration for: Integration process

Integrating SpotWave API for radio spot production is straightforward. After obtaining your API key from the dashboard, you can begin sending synthesis requests immediately. The endpoint mirrors standard TTS structures, allowing you to use existing OpenAI SDKs by simply adjusting the base URL to https://api.playcv99.com/v1.

API key setup

Generate your key via the /get-api-key/ page using Google or email. Use the header Authorization: Bearer tts-... for all requests. Note that issuing a new key immediately invalidates the old one, so manage your environment variables carefully to avoid mid-campaign interruptions.

  • Keys are single-active per account
  • Use Bearer tts-... format
  • New keys invalidate old ones instantly

Request structure

Send a POST request to /v1/audio/speech with a JSON body. Include model ("tts-1" or "tts-1-hd"), input (max 4096 characters), voice (e.g., "felix" for energetic promos), and speed (0.25–4.0). Omit language for English defaults, but specify ISO codes for other scripts to ensure accurate synthesis.

  • Endpoint: POST /v1/audio/speech
  • Max 4096 chars per request
  • Standard JSON body structure

Audio format selection

Choose response_format based on your post-production needs. MP3 is efficient for web delivery, while WAV and FLAC provide lossless quality. PCM is available for raw processing. Failed requests incur no charge, ensuring you only pay for usable assets.

  • MP3 for web/streaming
  • WAV/FLAC for broadcast masters
  • PCM for raw data

Quality control

Sound wave illustration for: Quality control

Quality control ensures your radio spots sound professional before broadcast. Since the API generates audio synchronously, you can verify the output instantly and discard any failed or unsatisfactory files without charge.

Use the speed parameter (0.25–4.0) to fine-tune pacing. Adjustments help match the voice to your specific ad length and brand tone, ensuring clear delivery within tight radio slots.

Listening for errors

Review the generated audio file for any artifacts or mispronunciations. Since requests are processed sequentially, ensure the synthesis completed successfully before assuming the audio is ready for use.

If a file sounds unnatural, check the input text for special characters or ambiguous punctuation that might confuse the engine.

  • Check for natural pauses and breath sounds
  • Verify correct pronunciation of brand names
  • Listen for clipping or distortion

Adjusting speed

Use the speed parameter to control the pacing of the narration. Values range from 0.25 to 4.0, allowing you to speed up dry reads or slow down complex explanations.

Optimal speeds often fall between 1.0 and 1.5 for standard radio spots, balancing clarity with energy.

  • Set speed to 1.5 for upbeat promos
  • Reduce to 0.8 for detailed instructions
  • Test extremes to find the sweet spot

Reviewing tone

Match the voice preset to your campaign’s mood. Use 'felix' for energetic promotions or 'aria' for warm, clear narration.

Ensure the voice aligns with your target demographic and the emotional tone of the script.

  • Use 'nova' for bright, upbeat ads
  • Choose 'sage' for instructional content
  • Select 'atlas' for neutral news-style reads

Licensing considerations

Sound wave illustration for: Licensing considerations

SpotWave API grants full commercial use rights for generated audio in radio spots and video advertisements. You can distribute the output files freely without recurring fees or usage caps tied to specific campaigns.

Commercial use rights

Generated audio is licensed for commercial use, including broadcast and digital advertising. There are no restrictions on the number of plays or impressions.

  • Full commercial license included
  • No per-play fees
  • Unlimited distribution

Attribution requirements

Attribution is not required for SpotWave audio, though the inaudible watermark identifies it as synthetic. You do not need to credit the voice actor or the API in your final radio spot or video content.

Watermark presence

Each generated file contains an inaudible watermark identifying it as synthetic speech. This does not affect audio quality or listener experience but serves as a proof of origin.

Licensing considerations

Common pitfalls

Sound wave illustration for: Common pitfalls

Radio spots demand precision, but small oversights can ruin production. Understanding these common pitfalls ensures your voice-overs sound professional and arrive on time without unexpected costs or delays.

Ignoring language codes

The API guesses language from the script, defaulting to English for Latin text. If you write a Dutch or French script without setting the language field, the engine may produce an unnatural accent or mispronounce names.

  • Explicitly set the language parameter for non-English scripts.
  • Verify output audio for accent accuracy before bulk generation.

Exceeding character limits

Each request accepts a maximum of 4,096 characters. Sending a long radio script in one call will fail. You must split longer scripts into segments, generate audio for each, and join the files.

Remember that spaces and punctuation count toward the 4,096-character limit. Plan your script length accordingly to avoid mid-sentence cuts.

Rate limiting issues

Your API key allows 60 requests per minute. If you trigger many parallel synthesis jobs, you may hit this limit. New accounts start with a lower limit of 10 requests per minute.

Implement retry logic with exponential backoff in your automation. This prevents errors during bulk processing of multiple ad variants.

Tool selection

Sound wave illustration for: Tool selection

Choosing the right tool depends on whether you prioritize immediate scalability or one-time control. API integration allows ad agencies to programmatically generate hundreds of radio spots daily, fitting seamlessly into existing content pipelines without manual file exports.

Software solutions often require local processing power and manual intervention for each file. For high-volume campaigns, the API’s predictable per-character billing and automation capabilities typically offer better operational efficiency than desktop alternatives.

API solutions generally scale better with volume, while software may incur higher upfront licensing costs. Our API charges $0.01 per 1,000 characters with no monthly fees, making it cost-effective for both sporadic and bulk production needs.

API vs. software

API integration allows ad agencies to programmatically generate hundreds of radio spots daily, fitting seamlessly into existing content pipelines without manual file exports.

Software solutions often require local processing power and manual intervention for each file. For high-volume campaigns, the API’s predictable per-character billing and automation capabilities typically offer better operational efficiency than desktop alternatives.

  • No local hardware constraints
  • Automated bulk synthesis
  • Direct cloud storage access

Cost comparison

Our API charges $0.01 per 1,000 characters with no monthly fees, making it cost-effective for both sporadic and bulk production needs.

Unlike subscription-based software, you only pay for what you use. Failed requests are free, and there are no hidden costs for processing time or file formats.

  • $0.01 per 1,000 characters
  • No monthly subscription
  • Pay-as-you-go model

Feature comparison

The API supports 8 distinct voice presets, 23 languages, and multiple audio formats including MP3, WAV, FLAC, OPUS, and PCM. It lacks SSML but offers precise speed control from 0.25x to 4.0x.

While it does not offer custom voice cloning or real-time streaming, the API provides reliable synthetic speech with inaudible watermarks for easy identification of generated content.

Use cases include generating voice-overs for video ads, radio spots, and podcast intros with consistent quality and tone across all generated files.

  • 8 voice presets
  • 23 languages supported
  • Multiple audio formats

Workflow optimization

Sound wave illustration for: Workflow optimization

Streamline your radio ad production by scripting voice-overs directly in code. Use the API’s speed parameter (0.25–4.0) to fine-tune pacing, ensuring the generated audio matches the precise timing requirements of 30-second or 60-second broadcast slots without manual audio editing.

Batch processing handles volume efficiently. Since requests are processed sequentially, split long scripts into 4096-character segments, synthesize each part, and join the resulting audio files. This approach ensures consistent character billing ($0.01 per 1,000 characters) and predictable delivery times for high-volume campaigns.

Automating scripts

Integrate the API into your content management system or ad-tech stack. Send JSON payloads with specific voice presets like 'felix' for energetic promos or 'aria' for warm narration. Automate the selection of the language ISO code to prevent default English guesses for non-English Latin script text, ensuring accurate pronunciation across your 23 supported languages.

  • Use Python or Node.js to wrap the POST /v1/audio/speech endpoint.
  • Map ad copy variations to specific voice IDs programmatically.
  • Set speed parameters to match exact second-counts required by radio stations.

Batch processing

For campaigns requiring multiple ad variants, process requests in order. The API handles up to 4096 characters per request; split longer scripts accordingly. Join the returned audio files to create complete commercials. Failed requests are free, so you can retry failed segments without incurring extra costs, ensuring 100% usable inventory.

  • Split scripts at natural pauses under 4096 characters.
  • Concatenate MP3 or WAV outputs sequentially.
  • Retry failed requests automatically; charges are refunded.

Quality assurance

Listen to generated clips for tone and pacing before final delivery. Adjust the speed parameter if the narration feels too rushed or slow. Remember that every file includes an inaudible watermark marking it as synthetic speech, which is standard for digital radio spots. Verify that the voice matches the intended emotional tone, such as 'atlas' for neutral news reads or 'nova' for upbeat product launches.

  • Check audio clarity and pacing against the original script.
  • Verify watermark presence for synthetic speech identification.
  • Ensure no SSML tags are used, as they are not supported.

Future trends

Sound wave illustration for: Future trends

The TTS landscape is shifting rapidly. While current models deliver high-quality synthetic speech, the industry is moving toward more nuanced emotional delivery and seamless multilingual capabilities within single voice profiles. This evolution promises richer audio for advertising without requiring complex voice design.

As demand for on-demand radio spots grows, the gap between batch processing and live generation narrows. Although true real-time streaming is not yet standard for bulk ad production, the infrastructure is aligning toward lower-latency delivery for dynamic content insertion.

AI voice evolution

Current models use licensed recordings from LibriTTS-R, ensuring clear, distinct voices. Future iterations may focus on more natural prosody and reduced robotic artifacts.

The industry is exploring finer control over pacing and emphasis without requiring SSML, allowing advertisers to tweak emotional tone directly via API parameters.

  • Improved naturalness in emotional delivery
  • Multilingual proficiency within single voices

Real-time generation

While current processing is batch-oriented, the API is optimizing for lower latency. This supports emerging use cases like dynamic radio ad insertion where ads are generated on-the-fly based on listener demographics.

Though not yet offering true WebSocket streaming, the gap between request and response is shrinking, enabling near-instant audio generation for short, targeted spots.

  • Lower latency for dynamic ads
  • Batch-to-stream hybrid workflows

Custom voice models

Currently, SpotWave offers eight presets without custom voice cloning. The industry trend is toward affordable, licensed voice cloning, but this requires explicit consent and legal frameworks. For now, advertisers rely on the existing eight voices, which cover most standard radio archetypes from warm narrators to energetic promo styles.

Custom models may eventually allow agencies to replicate brand ambassadors or specific talent, but current limitations mean sticking to the provided library.

  • Licensed voice cloning potential
  • Consent-driven model training

Support and resources

Sound wave illustration for: Support and resources

SpotWave API provides comprehensive documentation at /docs/ to guide your integration, covering all endpoints, parameters, and error codes. This resource is essential for developers setting up automated radio spot generation pipelines.

Documentation access

The documentation is available at /docs/ and details every endpoint, including request structures for audio synthesis and voice retrieval. It covers all parameters like speed, format, and model selection, ensuring you can build reliable batch processing workflows without guesswork.

  • Complete endpoint reference for /v1/audio/speech and /v1/voices
  • Code examples for Python, cURL, and JavaScript
  • Error code definitions and troubleshooting steps

Community forums

There are no official community forums for SpotWave API. Users rely on direct documentation and email support for specific technical inquiries.

Email support

Support is available exclusively via email. There is no phone number or live chat option. You can reach the team through the /contact/ page for issues regarding billing, key management, or integration questions.

Listen to the 8 voices

Each sample below was generated by this API from the same sentence.

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. Creative Commons Attribution 4.0
  3. TRON developer hub: TRC-20
  4. MDN: Web Speech API
  5. FFmpeg documentation
  6. LibriTTS-R paper (arXiv 2305.18802)
  7. Python documentation
  8. OpenAI API reference: create speech
  9. LibriTTS-R corpus (OpenSLR 141)
  10. ISO 639 language codes
  11. LibriVox public-domain audiobooks
  12. IETF RFC 7845: Ogg Opus

More on this site