Frequently asked questions
Who runs this site?
PandaTTS Review is an independent editorial resource for indie developers building car navigation and in-vehicle audio applications. We help indie developers choose the right text-to-speech engine for car navigation and in-vehicle audio apps. Our reviews focus on real-world usage scenarios like turn-by-turn instructions and long-form route narration, providing expert analysis based on latency, voice clarity, and cost.
Are the voices copyrighted?
Voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and public-domain audiobook readings from LibriVox. Each generated file carries an inaudible watermark marking it as synthetic speech. We do not offer custom voice cloning or use celebrity voices, ensuring clear licensing for your navigation audio applications.
Is my data used for training?
No. Input text is used only to produce the audio and is not used for training. Generated audio is returned in the response and not stored. We retain request metadata, such as time, characters, and status, strictly for billing purposes. Your data remains private and is not repurposed.
How long is data retained?
Input text is not stored after generation. Request metadata, including time, characters, and status, is kept for billing purposes. We do not retain the actual audio files or the original text input beyond the immediate generation process, ensuring your navigation route data remains transient and secure.
Can I contact you by phone?
No. We do not have a phone number for support. All communication is handled via email only. For indie developers needing assistance with API integration or billing questions, please reach out through our contact page. We do not offer live chat or direct phone lines for customer service.
Is there a live chat support?
No. We do not offer live chat support. Our support channel is email only. This allows us to maintain detailed records of your queries regarding latency, voice clarity, and pricing for your navigation apps. You can submit your questions through our contact page for a written response.
Are you affiliated with OpenAI?
No. We are not affiliated with OpenAI, ElevenLabs, Google, Amazon, Microsoft, or any other brand mentioned in comparisons. Our API endpoint mimics the OpenAI speech endpoint structure, so you can use the official OpenAI SDKs by changing the base_url. We are an independent provider of text-to-speech services.
Do you offer voice cloning?
No. We do not offer custom voice cloning for customers, nor do you upload your own voice. We provide eight preset voices based on licensed recordings. There is no voice design or emotion tagging available. Our focus is on providing reliable, clear synthetic speech for navigation and in-vehicle audio apps.
Is there an SLA for the API?
No. We do not provide an SLA or uptime guarantee. There is no enterprise contract or on-premise install option. Requests are processed one after another, so under load a request may wait in a queue. We offer a reliable API but do not guarantee specific availability percentages.
How are voices licensed?
Voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and public-domain audiobook readings from LibriVox. This ensures legal clarity for commercial use. Each generated file carries an inaudible watermark marking it as synthetic speech. We do not use celebrity voices or offer custom cloning, keeping licensing straightforward.
Is the audio stored after generation?
No. Generated audio is returned in the response and not stored by us. Input text is used only to produce the audio and is not used for training. We retain request metadata for billing, but the actual audio files and text inputs are not kept on our servers after delivery.
Do you have a privacy policy?
We maintain a privacy policy regarding data handling. Input text is used only to produce audio and is not used for training. Generated audio is not stored. Request metadata is kept for billing. We do not sell your data. Please refer to our site for full details on our privacy practices.
Can I request a data deletion?
Request metadata is kept for billing purposes. You can contact us via email to request deletion of your billing metadata. We do not store audio files or training data, so most of your data is already transient and not retained long-term.
Who validates the reviews?
PandaTTS Review is an independent editorial resource. We are an independent storefront, not affiliated with OpenAI or other brands. Our expert reviews focus on real-world usage scenarios like turn-by-turn instructions, comparing latency, voice clarity, and cost. We do not use sponsored content or external validation, relying on direct API testing.
Who We Are
PandaTTS Review is an independent editorial team dedicated to helping indie developers choose the right text-to-speech engine for car navigation and in-vehicle audio apps. We focus on practical metrics like latency, voice clarity, and cost, rather than marketing hype.
Our reviews are built on real-world usage scenarios, such as turn-by-turn instructions and long-form route narration, ensuring our data reflects actual developer needs.
Independent editorial team
We operate without corporate bias, evaluating the PandaTTS API strictly on its technical performance and pricing structure.
Focus on indie navigation developers
We tailor our analysis to the specific needs of small teams building car audio applications.
We prioritize solutions that offer clear voices for navigation cues without requiring complex infrastructure or high minimum spends.
Expert reviews with real data
Every claim in our reviews is backed by concrete numbers from live API requests.
We test character limits, speed settings, and voice options to provide accurate, actionable insights for your project.
Our comparisons focus on latency and clarity, the two factors that matter most for driving safety and user experience.
- Turn-by-turn instruction clarity
- Long-form route narration quality
- Real-time latency under load
Voice Licensing and Sources
PandaTTS Review verifies the legal and technical foundations of our recommended engine. We prioritize clarity and compliance for developers who need reliable, licensable audio assets for commercial in-vehicle applications without hidden royalty fees.
LibriTTS-R license (CC BY 4.0)
Our TTS engine uses voices derived from the LibriTTS-R dataset, released under the Creative Commons Attribution 4.0 International license. This permits commercial use provided proper attribution is given.
- CC BY 4.0 allows commercial integration
- Clear attribution requirements for compliance
- Standardized license for predictable costs
Recordings from LibriVox
The underlying audio samples originate from LibriVox, a platform hosting volunteer-read public domain audiobooks. These recordings provide the natural speech patterns and prosody that our synthesis models replicate.
- Volunteer-read source material
- Natural prosody for navigation cues
- High-quality baseline audio data
Public domain audiobook readings
By leveraging public domain readings, we ensure the base voices are free from complex third-party licensing constraints. This allows us to maintain transparent pricing and stable voice availability without renegotiating rights for individual audio segments.
- No recurring royalty fees
- Stable voice availability
- Transparent licensing structure
Data Policy and Privacy
We process your input text strictly to generate the requested audio file. The text is not retained for any other purpose once the synthesis is complete. This ensures that your navigation instructions or route narration content are not repurposed beyond immediate audio generation.
Our system does not use your input text to train or improve our underlying speech models. Your data is used solely for the specific request you send via the API, ensuring a clear boundary between your proprietary navigation scripts and our general engine improvements.
Input text used only for audio
When you send text to the API, it is processed exclusively to create the audio response. We do not archive or repurpose your input strings for marketing, analysis, or other services.
This approach keeps your navigation logic and route data distinct from our training datasets.
- Text is consumed only during synthesis
- No secondary use of input strings
- Data remains isolated per request
Not used for training models
Your input text is not used to train or improve our underlying speech models. Your data is used solely for the specific request you send via the API, ensuring a clear boundary between your proprietary navigation scripts and our general engine improvements.
This ensures that your specific phrasing or unique navigation instructions do not influence future model behavior.
- Input text excluded from training sets
- No influence on model evolution
- Predictable output behavior
Generated audio not stored
The generated audio files are returned in the response and are not stored on our servers after delivery. This minimizes data retention and reduces the risk of your navigation audio being accessed later.
While request metadata (time, characters, status) is kept for billing purposes, the actual audio content itself is transient, ensuring your generated voiceovers remain ephemeral unless you save them yourself.
This transient storage aligns with the needs of indie developers who manage their own audio assets and prefer minimal data footprints on our side.
- Audio returned in response
- No server-side audio retention
- Minimal data footprint
Billing and Metadata Retention
PandaTTS Review ensures you know exactly what you pay for. We retain request metadata—specifically the timestamp, character count, and processing status—to provide a clear audit trail for every billing cycle. This data is used solely for accurate charge calculation and dispute resolution, not for training our models or analyzing content.
Each request’s metadata is logged to verify that you are charged only for successfully processed characters. Failed requests are excluded from billing, ensuring you never pay for wasted processing power. This transparent tracking allows indie developers to monitor their API usage precisely without hidden fees or ambiguous charges.
Request metadata kept for billing
We store the metadata associated with each TTS request to ensure accurate billing. This includes the time of the request, the input length, and whether the synthesis succeeded. This information is retained only for the purpose of verifying charges and resolving any billing discrepancies.
By keeping this data, we guarantee that you are billed correctly for every character processed. If a request fails, the metadata confirms that no charge should apply, protecting your budget from unnecessary costs.
- Timestamp of each request
- Input character count
- Synthesis status (success/fail)
Time, characters, and status logged
Every interaction is logged with precise timing and character metrics. This allows us to verify that you are charged exactly $0.01 per 1,000 characters, with spaces and punctuation included. The status log confirms whether the audio was generated successfully.
This logging ensures transparency. You can see exactly when your requests were processed and how many characters were consumed, providing a clear record of your API usage over time.
Transparent usage tracking
Our billing system provides a clear view of your consumption. You can see exactly how many characters you have used and how much credit remains in your balance. This tracking is essential for indie developers managing tight budgets.
We do not hide usage details. The metadata we retain is directly tied to your billing statement, ensuring that every charge is justified and verifiable. This transparency helps you plan your app’s audio costs accurately.

Limitations and Honest Claims
PandaTTS Review provides an honest assessment of the PandaTTS API for car navigation developers. We focus on practical metrics like latency and cost, avoiding marketing hype. Our reviews are based on real-world usage scenarios, such as turn-by-turn instructions and long-form route narration, ensuring you get data-driven insights.
No SLA or uptime guarantee
We do not promise 99.9% uptime or define service level agreements. The API processes requests sequentially, so under heavy load, synthesis may wait in a queue. This is typical for our infrastructure and affects real-time navigation apps differently than batch processing.
No custom voice cloning
The API offers eight preset voices derived from licensed recordings. We do not offer custom voice cloning, where you upload a sample to create a unique voice. If your navigation app requires a specific brand voice, you must select from our existing library.
No official brand affiliation
PandaTTS Review is an independent editorial team. We are not affiliated with OpenAI, ElevenLabs, Google, Amazon, or Microsoft. Our comparisons are based on published facts and real testing, not promotional partnerships. We help indie developers choose the right engine without brand bias.
Contact and Support
PandaTTS Review maintains a lean, developer-focused support structure to keep overhead low and transparency high. We prioritize clear, asynchronous communication that respects the busy schedules of indie developers building car navigation and in-vehicle audio apps.
All inquiries are handled through our dedicated support email. We do not maintain a phone line or live chat widget, ensuring every question is logged and traceable for technical reference.
Email support only
Send your questions to our contact page. Whether you need clarification on voice licensing, billing discrepancies, or technical integration details, our editorial team responds directly to your inbox.
No phone number or live chat
We do not offer a phone number or real-time live chat. This deliberate choice allows us to provide detailed, written responses that you can reference later during your development workflow without waiting on hold.
Response time expectations
We aim to acknowledge all emails within one business day. Complex technical queries may take slightly longer as we verify details against our usage logs and billing records.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. If your app targets users who value aesthetic experience alongside navigation accuracy, these presets offer a distinct advantage.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- openai-python on GitHub
- LibriTTS-R paper (arXiv 2305.18802)
- LibriTTS-R corpus (OpenSLR 141)
- Node.js documentation
- ISO 639 language codes
- Creative Commons Attribution 4.0
- curl documentation
- LibriVox public-domain audiobooks
- W3C WAI: making audio and video accessible
- FFmpeg documentation
- IETF RFC 7845: Ogg Opus
