FAQ
FAQ
Product
How many languages do you support?
We support 9 languages and 14 locales including English (US, UK, Australia, India, Canada), Spanish (Spain, Mexico), Chinese (Mandarin), Korean, French, German, Japanese, Italian, and Polish. We are adding 25 more languages.
What evaluation types & criteria do you support?
We provide wide-used evaluation types through templates including voice naturalness, quality, comparison, deep quality analysis, speaker similarity, preferences, and more. You can even customize your own evaluations.
How can I evaluate my voice AI in a language that I don’t know?
We can provide a text corpus for your output generation.
I don’t know how to evaluate. Where to start?
You can take a look at this product video (https://go.podonos.com/demo) and the docs (https://www.podonos.com/docs). Feel free to schedule a call with us at https://go.podonos.com/book-a-demo.
Can you detect emotions of voice output?
We have an API (in beta) for emotion detection including confident, passionate, relaxed, pleasant, calm, tired, surprised, anxious, annoyed, stressed, nervous, and many more.
How can I evaluate the naturalness of cloned voices?
You can measure a voice similarity between two audio sample pairs. See https://www.podonos.com/docs/similarity
Can you help us compare with other existing voice APIs?
Not yet, but it is under our radar. Let us know.
Do you also measure the API latency?
Not yet, but it is under our radar. Let us know.
We found mispronunciations, low quality voices, or bad accents. How can we fix them?
If you know how to train or refine your AI model, we are happy to provide a diverse and large set of data for your model refinement. Please schedule a call in this link (https://go.podonos.com/book-a-demo)
I have a voice call between an AI agent and customers. Do you evaluate and analyze them?
Yes.
How long does it take to run an evaluation?
Whether the size is big or small, it is less than 12 hours. As of 2025 Q1, 80% of evaluations take less than 4 hours.
What is the ideal audio length?
The optimal length of each audio for evaluation is 5-15 seconds.
How many languages do you support?
We support 9 languages and 14 locales including English (US, UK, Australia, India, Canada), Spanish (Spain, Mexico), Chinese (Mandarin), Korean, French, German, Japanese, Italian, and Polish. We are adding 25 more languages.
What evaluation types & criteria do you support?
We provide wide-used evaluation types through templates including voice naturalness, quality, comparison, deep quality analysis, speaker similarity, preferences, and more. You can even customize your own evaluations.
How can I evaluate my voice AI in a language that I don’t know?
We can provide a text corpus for your output generation.
I don’t know how to evaluate. Where to start?
You can take a look at this product video (https://go.podonos.com/demo) and the docs (https://www.podonos.com/docs). Feel free to schedule a call with us at https://go.podonos.com/book-a-demo.
Can you detect emotions of voice output?
We have an API (in beta) for emotion detection including confident, passionate, relaxed, pleasant, calm, tired, surprised, anxious, annoyed, stressed, nervous, and many more.
How can I evaluate the naturalness of cloned voices?
You can measure a voice similarity between two audio sample pairs. See https://www.podonos.com/docs/similarity
Can you help us compare with other existing voice APIs?
Not yet, but it is under our radar. Let us know.
Do you also measure the API latency?
Not yet, but it is under our radar. Let us know.
We found mispronunciations, low quality voices, or bad accents. How can we fix them?
If you know how to train or refine your AI model, we are happy to provide a diverse and large set of data for your model refinement. Please schedule a call in this link (https://go.podonos.com/book-a-demo)
I have a voice call between an AI agent and customers. Do you evaluate and analyze them?
Yes.
How long does it take to run an evaluation?
Whether the size is big or small, it is less than 12 hours. As of 2025 Q1, 80% of evaluations take less than 4 hours.
What is the ideal audio length?
The optimal length of each audio for evaluation is 5-15 seconds.
Evaluators
How many are there? Who are they?
Our service is connected with 150,000 human evaluators with qualified intelligibility level and good hearing capability in a quiet place with headphones.
Are they qualified for deep voice analysis?
We qualify their identity, hearing capability, and language skills. Every time they join an evaluation session, we confirm that they wear proper headphones in a quiet environment.
Do you work with domain experts?
We would like to hear more about the domains. Please schedule a chat in https://go.podonos.com/book-a-demo
How many are there? Who are they?
Our service is connected with 150,000 human evaluators with qualified intelligibility level and good hearing capability in a quiet place with headphones.
Are they qualified for deep voice analysis?
We qualify their identity, hearing capability, and language skills. Every time they join an evaluation session, we confirm that they wear proper headphones in a quiet environment.
Do you work with domain experts?
We would like to hear more about the domains. Please schedule a chat in https://go.podonos.com/book-a-demo
Pricing
How much does it cost?
The maximum cost is $0.22 per rating. For additional discounts, see [pricing].
What is a recurring deposit?
It is a monthly recurring deposit. Your deposit doesn’t disappear. You can use the deposit anytime. You may use all the deposits in a single evaluation. You can even pull the deposit back without any fee.
I am a student. Can you provide a discount?
Yes if you are actively registered as a student. We will provide an additional 5% discount. Please contact us.
I am a small startup. Can you provide a discount?
Yes if your company has received less than $2MM total funding or less than $1M annual revenue . We will provide an additional 5% discount. Please contact us.
How much does it cost?
The maximum cost is $0.22 per rating. For additional discounts, see [pricing].
What is a recurring deposit?
It is a monthly recurring deposit. Your deposit doesn’t disappear. You can use the deposit anytime. You may use all the deposits in a single evaluation. You can even pull the deposit back without any fee.
I am a student. Can you provide a discount?
Yes if you are actively registered as a student. We will provide an additional 5% discount. Please contact us.
I am a small startup. Can you provide a discount?
Yes if your company has received less than $2MM total funding or less than $1M annual revenue . We will provide an additional 5% discount. Please contact us.
Security
Is Podonos secure?
Yes. We securely keep your data and files using AES-256 encryption. Your data is used only for running evaluation. Optionally your data may be in public only when you share the evaluation result with everyone in the world.
I uploaded my audio files. Where are they stored?
All your data is stored in the United States. See this : https://podonos.com/subprocessors
Can a human evaluator download all?
No. We automatically split the session. Also, downloading data is blocked and prohibited. Every data is decrypted right before being played on the web browser.
Is Podonos secure?
Yes. We securely keep your data and files using AES-256 encryption. Your data is used only for running evaluation. Optionally your data may be in public only when you share the evaluation result with everyone in the world.
I uploaded my audio files. Where are they stored?
All your data is stored in the United States. See this : https://podonos.com/subprocessors
Can a human evaluator download all?
No. We automatically split the session. Also, downloading data is blocked and prohibited. Every data is decrypted right before being played on the web browser.
Others
I have a feature request or a question
Please feel free to contact us at hello@podonos.com. Our CEO will directly respond to your question typically in less than 12 hours.
I have a feature request or a question
Please feel free to contact us at hello@podonos.com. Our CEO will directly respond to your question typically in less than 12 hours.
Ready to unlock the potential of your voice AI Model?
Ready to unlock the potential of your voice AI Model?
Improve your model with trust
Improve your model with trust