Text to Voice Over API for Scalable Dubbing Pipelines

By SyncVoice editorial team · Updated 2026-11-18

This page details the SyncVoice API, a programmatic text-to-speech solution built for localization agencies processing high volumes of video voice-overs. We provide a reliable infrastructure that replaces manual recording workflows with scalable audio generation, ensuring consistent voice quality across large-scale dubbing projects.

You will find specific technical specifications, integration examples for Python and cURL, and a transparent pricing model based on character usage. This resource is designed for technical leads and operations managers who need to integrate TTS into their existing pipelines without managing complex vendor contracts or subscription overhead.

Text to Voice Over API for Scalable Dubbing Pipelines

Frequently asked questions

What is a text to voice over API?

SyncVoice is a cloud text-to-speech API that accepts text via POST requests and returns audio files. It uses a JSON body similar to the OpenAI speech endpoint, allowing localization agencies to integrate programmatic audio generation directly into their existing dubbing pipelines without managing manual recording workflows.

How does TTS compare to human voice actors for dubbing?

TTS replaces manual recording workflows with scalable programmatic generation at roughly real-time speed. While human actors provide unique nuance, SyncVoice offers consistent quality across 23 languages and 8 presets. It eliminates scheduling logistics and allows instant iteration, though it lacks the emotional range of human performance.

Which voices are best for documentary narration?

For documentary narration, Orion is the designated choice, described as deep and documentary-style. Atlas provides a neutral, news-oriented tone suitable for factual content. Both are available in multiple languages, ensuring consistent audio quality for localized documentary projects without requiring specific voice actor hiring.

How many languages does the API support?

The API supports 23 languages: Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. Any voice can speak any of these languages, and the language can be specified via ISO code.

Is there a limit on text length per request?

Yes, each request accepts up to 4096 characters of input text. For longer scripts, you must split the text into parts of this maximum length and join the resulting audio files. Spaces and punctuation count toward this character limit, so plan your chunking accordingly.

How is the price calculated per character?

Pricing is $10 per 1 million characters, or $0.01 per 1,000 characters. Spaces and punctuation count as characters. Failed requests are free, with the charge returned in full. There are no monthly plans, seats, or minimum spend requirements, allowing you to pay only for what you generate.

Can I use the audio files commercially?

Yes, the API is designed for commercial use by localization agencies processing high volumes of video voice-overs. The generated audio is returned in the response and not stored, allowing you to integrate the files directly into your production pipeline without ongoing licensing fees or subscription constraints.

Do I need to credit the voice actors?

The voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobook readings. The voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobook readings.

What happens if a synthesis request fails?

If synthesis fails, the charge is returned in full. You are not billed for failed requests. This ensures that your budget is only consumed by successfully generated audio files, protecting your costs during high-volume localization workflows where occasional processing errors might occur under load.

Is the API compatible with OpenAI SDKs?

Yes, the request shape is the same as the OpenAI speech endpoint. You can use official OpenAI SDKs by changing the base_url to https://api.playbd2.com/v1 and using your SyncVoice API key. This allows for easy integration into existing codebases that already support OpenAI-style TTS endpoints.

How long does it take to generate audio?

Typical speed is roughly real time, generating about 25 characters of text per second of audio. Short phrases may take a few seconds, while longer texts process at real-time speed. Requests are processed sequentially, so under heavy load, a request may wait in a queue before generation begins.

Can I control the speed of the speech?

Yes, you can control speech speed using the speed parameter, which accepts values between 0.25 and 4.0. This allows you to adjust the pacing of the narration to match the visual timing of your video content precisely, ensuring seamless synchronization between the audio and the video track.

What audio formats are available for output?

The API supports MP3, WAV, OPUS, FLAC, and PCM output formats. You can specify the desired format in the response_format field of your request. This flexibility allows you to choose the appropriate codec for your post-production workflow, whether you need lossless quality or compressed efficiency.

How do I sign up for the API key?

Sign up on the /get-api-key/ page using Google or an email and password. Accounts are per site. After signing up, you receive a Bearer token key. Google sign-ins include a $0.50 trial credit valid for 7 days, while email accounts start without trial credit.

First request

curl

curl https://api.playbd2.com/v1/audio/speech \
  -H "Authorization: Bearer $TTS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"tts-1","voice":"aria","input":"Hello from the API.","response_format":"mp3"}' \
  --output hello.mp3

Automate Your Dubbing Workflow

Sound wave illustration for: Automate Your Dubbing Workflow

SyncVoice replaces manual recording workflows with programmatic audio generation. Send text via API and receive audio files directly into your localization pipeline, eliminating the logistical overhead of scheduling voice actors and studio time for every project.

Our system ensures consistent voice quality across all generated content. Each voice profile maintains uniform tone and pacing, allowing localization agencies to standardize output for multiple languages without varying performance quality between different talent pools.

Programmatic Audio Generation

Programmatic Audio Generation

Integrate TTS directly into your CMS or DOP systems. Use standard HTTP endpoints to trigger synthesis on demand, processing up to 4096 characters per request. Handle long scripts by splitting text programmatically and joining the resulting audio segments seamlessly.

  • Standard REST API endpoint
  • JSON request body
  • Automated text splitting for long scripts

Consistent Voice Quality

Maintain brand voice across all localized videos. Our eight preset voices deliver uniform tone, ensuring that a character’s audio profile remains identical whether generated for English, Spanish, or Japanese versions of your content.

No more variations between different voice actors. Each generation uses the same engine and voice preset, guaranteeing predictable audio characteristics for every video in your catalog.

  • Uniform tone across 23 languages
  • Same voice, multiple markets
  • Predictable audio characteristics

Integration with Existing Tools

Use the official OpenAI SDKs with a simple base_url change. Integrate with your existing video editing software or dubbing management platforms using standard HTTP requests. No proprietary plugins or custom middleware required for basic integration.

Generate audio files in MP3, WAV, OPUS, FLAC, or PCM formats. Compatible with most professional video editing suites and streaming platforms, ensuring your generated voice-overs fit directly into your existing post-production workflow.

  • OpenAI SDK compatible
  • Standard HTTP requests
  • Multiple audio formats

Licensing and Voice Assets

Sound wave illustration for: Licensing and Voice Assets

SyncVoice generates audio from licensed recordings sourced from the LibriTTS-R dataset (CC BY 4.0), originally derived from public-domain LibriVox audiobooks. This foundation ensures consistent, high-quality synthetic speech suitable for professional localization pipelines without complex voice design overhead.

Every generated audio file contains an inaudible watermark identifying it as synthetic speech, providing clear attribution for compliance. The system offers eight distinct voice presets across male and female profiles, allowing teams to select appropriate tones for diverse content types.

LibriTTS-R Licensed Recordings

Our voice models are built on the LibriTTS-R dataset, which utilizes recordings from LibriVox public-domain audiobook readings. This licensing structure provides a clear legal foundation for commercial voice-over generation, avoiding the ambiguity often associated with raw neural network training data.

Inaudible Synthetic Watermark

Each audio file generated via the API includes an inaudible watermark that marks the content as synthetic speech. This feature aids in content identification and compliance tracking within large-scale dubbing workflows where distinguishing between human and AI-generated audio is necessary.

8 Preset Voice Profiles

We provide eight curated voice presets designed for different narration styles, ranging from warm narrators to deep documentary tones. These voices are optimized for clarity and consistency, ensuring uniform audio quality across long-form video projects.

  • Aria: Warm, clear narrator
  • Nova: Bright, upbeat
  • Luna: Soft, storytelling
  • Sage: Calm, instructional
  • Felix: Energetic, promo
  • Milo: Friendly, conversational
  • Atlas: Neutral, news
  • Orion: Deep, documentary

Supported Languages and Formats

Sound wave illustration for: Supported Languages and Formats

SyncVoice API supports 23 languages, covering major markets for localization agencies including English, Spanish, German, French, Japanese, and Mandarin Chinese. Each voice can deliver clear, native-quality audio in any of these languages, ensuring consistent branding across diverse regional releases.

The API returns audio in MP3, WAV, OPUS, FLAC, or PCM formats. This flexibility allows your pipeline to integrate directly with existing video editing and dubbing tools without requiring intermediate conversion steps, reducing processing overhead for high-volume projects.

23 Languages Supported

Support includes Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. If a required language is not listed, it is not supported.

  • Arabic, Chinese, Danish, Dutch, English
  • Finnish, French, German, Greek, Hebrew
  • Hindi, Italian, Japanese, Korean, Malay
  • Norwegian, Polish, Portuguese, Russian, Spanish
  • Swahili, Swedish, Turkish

Multiple Audio Formats (MP3, WAV, OPUS, FLAC)

Select the output format via the response_format parameter. MP3 and OPUS are compressed formats suitable for web delivery and storage efficiency. WAV and FLAC provide lossless quality suitable for final master archiving. PCM is available for raw audio processing.

All generated files include an inaudible watermark identifying them as synthetic speech, aiding in content verification and rights management.

  • MP3: Standard compressed format
  • WAV: Lossless uncompressed audio
  • OPUS: Efficient web streaming
  • FLAC: Lossless archival quality
  • PCM: Raw audio data

Language Auto-Detection

Set the optional language field to an ISO code for precise control. If omitted, the system guesses the language from the script. Latin script defaults to English; non-English Latin text requires explicit language setting.

This ensures accurate pronunciation and intonation for languages like German, Dutch, or Japanese without manual configuration for every batch.

Technical Specifications

Sound wave illustration for: Technical Specifications

SyncVoice API provides a standard HTTP REST endpoint compatible with existing OpenAI SDKs. By adjusting the base_url to https://api.playbd2.com/v1, teams can utilize familiar libraries for seamless integration into their localization pipelines without rewriting core logic.

OpenAI SDK Compatible

The API mirrors the OpenAI speech endpoint structure, allowing immediate use of official OpenAI SDKs. Only the base_url and authentication header require modification, minimizing development effort for engineering teams.

  • Compatible with official OpenAI SDKs
  • Standard JSON request body
  • Familiar tts-1 and tts-1-hd model identifiers

Standard HTTP REST Endpoint

Audio generation is handled via a simple POST request to /v1/audio/speech. The system returns audio files directly, supporting formats like MP3, WAV, OPUS, FLAC, and PCM for flexible downstream processing.

  • Endpoint: POST /v1/audio/speech
  • JSON input body
  • Direct audio file response

Rate Limits and Quotas

Each API key is limited to 60 requests per minute, with an initial limit of 10 requests per minute before the first top-up. This ensures stable performance for high-volume batch processing.

  • 60 requests/minute per key
  • 10 requests/minute initial limit
  • Automatic rate enforcement

Pricing Model

Sound wave illustration for: Pricing Model

SyncVoice charges $0.01 per 1,000 characters, with spaces and punctuation included. Failed requests are free, ensuring you pay only for successful synthesis. The free trial provides $0.50 credit for 7 days upon Google sign-in.

Pay Per Character Usage

Billing is calculated at $10 per million characters. This straightforward metric simplifies budgeting for large localization projects without hidden fees for specific voices or formats.

No Monthly Subscriptions

There are no monthly plans, seats, or minimum spend requirements. You top up via USDT (TRC-20) on the TRON network, with a $5 minimum. Balances do not expire, allowing flexible usage across teams.

Transparent Pricing

All costs are predictable based on character count. Rate limits apply (60 requests/minute), but do not affect pricing. Use the dashboard to track usage and balance in real-time.

  • Top-up via USDT (TRC-20)
  • Balance never expires
  • No VAT or invoices
Pricing Model

Data Privacy and Security

Sound wave illustration for: Data Privacy and Security

SyncVoice API ensures that input text is processed solely for audio synthesis and is never retained or used to train our underlying models. This guarantees that your proprietary scripts and localization assets remain confidential and are not repurposed for model improvement.

Generated audio files are returned directly in the API response and are not stored on our servers after delivery. Only request metadata—such as timestamp, character count, and status—is retained for billing purposes, ensuring a minimal footprint for audit compliance.

Input Text Not Used for Training

Unlike some consumer-grade services, SyncVoice does not use your submitted text to enhance or train the TTS engine. Your localization content remains strictly within the scope of the immediate generation request.

No Long-Term Audio Storage

The synthesized audio files are delivered via the response stream and are not archived on our infrastructure. This reduces liability and storage costs, allowing your team to manage audio assets directly within your own content management or dubbing pipelines.

Metadata Retention for Billing

We retain only the necessary metadata required for accurate billing: request time, character count, and processing status. This data is used exclusively for invoice generation and usage tracking, ensuring transparency without unnecessary data retention.

Voice Profiles Overview

Sound wave illustration for: Voice Profiles Overview

SyncVoice provides eight distinct voice profiles, derived from licensed LibriTTS-R recordings and public-domain audiobook readings. These voices are designed for consistent tone across localized video content, ensuring that a single character maintains the same vocal identity throughout a series.

Female Narrators (Aria, Nova, Luna, Sage)

Aria offers a warm, clear narrator tone suitable for general narration. Nova provides a bright, upbeat voice for energetic content. Luna delivers a soft, storytelling quality ideal for narrative-driven projects. Sage ensures a calm, instructional presence for educational or explanatory videos.

  • Aria: Warm, clear narrator
  • Nova: Bright, upbeat
  • Luna: Soft, storytelling
  • Sage: Calm, instructional

Male Narrators (Felix, Milo, Atlas, Orion)

Felix brings an energetic, promotional style to commercial voice-overs. Milo offers a friendly, conversational tone for direct-to-camera content. Atlas provides a neutral, news-style delivery for factual reporting. Orion delivers a deep, documentary voice for serious, long-form content.

  • Felix: Energetic, promo
  • Milo: Friendly, conversational
  • Atlas: Neutral, news
  • Orion: Deep, documentary

Tone and Style Variations

Each voice is a preset based on specific licensed recordings. They cannot be cloned or customized with emotion tags. The tone is fixed per voice, allowing teams to select the appropriate profile for each character or segment in their localization pipeline.

Integration Examples

Sound wave illustration for: Integration Examples

SyncVoice API maintains compatibility with the official OpenAI Python SDK, requiring only a base URL adjustment to https://api.playbd2.com/v1. This allows localization teams to deploy existing automation scripts with minimal code changes, ensuring rapid integration into current CI/CD pipelines for video dubbing workflows.

The standard HTTP REST endpoint accepts JSON payloads, making it accessible via any programming language or tool that supports basic HTTP requests. This universality ensures that legacy systems and modern microservices can both utilize the TTS engine without custom middleware.

Python SDK Configuration

Initialize the client by setting the base URL and providing your Bearer token. The SDK handles request serialization and response parsing automatically, returning the audio binary directly for file storage.

Ensure the base_url points to https://api.playbd2.com/v1 and use the standard openai package. This approach leverages established client libraries for connection pooling and error handling within your automation framework.

  • Set base_url to https://api.playbd2.com/v1
  • Use Authorization: Bearer tts-<your_key> header
  • Standard OpenAI SDK compatibility

cURL Command Examples

Direct HTTP requests allow for quick testing and integration with shell scripts or cron jobs. Send a POST request to the speech endpoint with the required JSON body containing the model, input text, and voice selection.

The response returns the audio file directly, which can be piped to a local file or forwarded to a storage bucket. This method is ideal for simple automation tasks or debugging specific voice outputs before full SDK deployment.

  • Endpoint: POST /v1/audio/speech
  • Header: Authorization: Bearer tts-<key>
  • Body: JSON with model, input, voice

Batch Processing Logic

For high-volume localization, split input text into chunks of up to 4096 characters. Generate audio for each chunk independently to manage memory usage and allow parallel processing across multiple API keys or threads.

Concatenate the resulting audio files in sequence to reconstruct the full voice-over. This method ensures consistent voice quality while adhering to the character limit per request, optimizing throughput for large video scripts.

  • Split text at 4096 character boundaries
  • Process chunks in parallel
  • Concatenate audio segments sequentially

Why Agencies Choose SyncVoice

Sound wave illustration for: Why Agencies Choose SyncVoice

SyncVoice eliminates the overhead of coordinating multiple freelance voice actors for large-scale localization projects. By generating audio programmatically, agencies maintain consistent quality across thousands of hours of content without managing individual contracts or scheduling conflicts.

Cost Efficiency for Large Volumes

Pricing is strictly $10 per million characters, ensuring predictable costs for high-volume dubbing pipelines. Failed requests incur no charge, protecting your budget during peak processing periods. With no monthly subscriptions or minimum spend requirements, you only pay for the audio you actually receive.

  • $0.01 per 1,000 characters including spaces
  • No monthly base fees or seat limits
  • Full refund on synthesis failures

Rapid Turnaround Times

Standard processing generates audio at roughly real-time speed, allowing you to scale output without waiting for human availability. For urgent localization deadlines, the API processes requests sequentially, ensuring consistent delivery times even under load.

  • Approx. 25 characters generated per second
  • Scalable without hiring additional talent
  • Immediate audio delivery via REST endpoint

Simplified Vendor Management

Replace complex vendor negotiations with a single API key. Manage all voice assets, languages, and formats through a unified dashboard. Support is available via email for technical issues, reducing the administrative burden of managing multiple external providers.

  • Single point of contact for all audio needs
  • Unified management of 8 voices and 23 languages
  • Email support for technical queries

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. This scope targets core localization needs without unnecessary fragmentation.

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. curl documentation
  4. W3C WAI: making audio and video accessible
  5. IETF RFC 7845: Ogg Opus
  6. LibriVox public-domain audiobooks
  7. LibriTTS-R corpus (OpenSLR 141)
  8. LibriTTS-R paper (arXiv 2305.18802)
  9. FFmpeg documentation
  10. OpenAI API reference: create speech
  11. Python documentation
  12. Node.js documentation

More on this site