FAQ
Straight answers before you send an email
Pricing, pilots, confidentiality, quality measurement and the questions clients ask most often about each service.
Working with us
- What does NewAgeSolution do?
- NewAgeSolution is a translation, localization and data services company. We collect, annotate, translate and quality check image, text, audio, video and facial data for product and research teams, and we localize products entering new markets.
- Which services do you offer?
- Image data collection and annotation, text data and language annotation, speech collection and transcription, video capture and annotation, facial data with recorded consent, and translation and localization.
- How do you price a project?
- Pricing depends on volume, annotation complexity, languages and turnaround. We quote per unit for repeatable work and per project for scoped pilots, and we share the rate breakdown before work starts.
- What is the smallest project you take on?
- We run paid pilots from a few thousand items. A pilot lets both sides agree guidelines and quality targets before committing to a larger volume.
- How long does a project take?
- A pilot usually runs one to two weeks including guideline work. Ongoing delivery is scheduled against an agreed weekly volume, and we confirm the timeline in writing before starting.
- How do you protect confidential data?
- Work runs under a signed agreement with access limited to assigned team members. We support working inside your own tooling and storage when the data cannot leave your products.
- Where are your annotators based?
- Our core team is based in India with a contributor network across several regions and languages, which lets us staff projects with native speakers of the target language.
- Can you work inside our annotation platform?
- Yes. We work in your platform when you have one, and we provide tooling when you do not. Either way you receive the same guideline document and quality reporting.
- How do you measure quality?
- We sample each batch, score it against the guideline, and report accuracy and annotator agreement with delivery. Batches below the agreed threshold are reworked before they reach you.
- How do I start a project?
- Send a note to connect@newagesolution.in describing the data type, languages and rough volume you need. We reply with questions, a suggested pilot scope and a quote.
Image data collection and annotation
- What image annotation formats do you deliver?
- We deliver COCO JSON, YOLO text files, Pascal VOC XML, CSV or a custom schema you define. We match your internal workflow rather than asking you to convert files.
- Can you collect images in a specific country or setting?
- Yes. Our contributor network lets us commission fresh capture by region, device type, lighting and scene, with consent recorded for every submission.
Text data and language annotation
- Do you write instruction and comparison data?
- Yes. Trained writers produce sample instructions, responses and ranked comparisons against your policy document, with a second reviewer checking every item before delivery.
- Which languages do you cover for text work?
- English plus Hindi, Marathi, Bengali, Tamil, Telugu, Kannada, Gujarati, Punjabi, Malayalam, Urdu and several others, including mixed script writing common in Indian messaging.
Speech collection and transcription
- How accurate are your transcripts?
- Delivery targets 99 percent accuracy on clear speech. Each file passes a transcription pass and an independent review pass, and we report the sampled error rate with the batch.
- Can you record speakers in noisy real settings?
- Yes. We commission capture in homes, vehicles, shops and streets when the product needs to work in those conditions, and we tag the noise types present.
Video capture and annotation
- Do you handle long recordings?
- Yes. Long files are split into reviewable segments with identity carried across cuts, so tracks stay consistent when the segments are joined again.
- Can you film scenarios that rarely occur naturally?
- Yes. Rare events are staged safely with briefed participants so the product sees the situations it would otherwise almost never encounter in collected footage.
Facial data with recorded consent
- How do you handle consent for facial data?
- Contributors read a plain language release that names the purpose, retention period and withdrawal process, then sign before capture. Records are retained and can be produced during an audit.
- Can a contributor withdraw their facial data later?
- Yes. A withdrawal request removes the contributor from our active store and we notify the client so downstream copies can be removed under the terms of the project agreement.
Translation and localization
- Which languages do you translate into?
- Major Indian languages including Hindi, Marathi, Bengali, Tamil, Telugu, Kannada, Gujarati, Punjabi, Malayalam and Urdu, plus Spanish, French, German, Portuguese, Arabic, Indonesian, Japanese and more on request.
- What is the difference between translation and localization?
- Translation converts the words. Localization also adapts currency, dates, measurements, names, images, tone and legal wording so the content fits local expectations and reads naturally.
- Do you post edit pre translated content?
- Yes. We offer light post editing for internal or high volume content and full post editing for anything customer facing, performed by native speakers of the target language.
- Do you work with translation memory and CAT tools?
- Yes. We work in Trados, memoQ, Phrase, Crowdin, Lokalise and similar platforms, or in your own system. The translation memory and glossary belong to you and are returned with delivery.
- How fast can you turn around a translation?
- A standard working day covers around 2,000 words per linguist per language, with review on top. Larger volumes are split across a team with a shared glossary so wording stays consistent.
- How do you keep quality consistent across languages?
- Every language runs with a native translator and a separate native reviewer, against one glossary and style guide. Review findings are scored by error type so problems can be traced and fixed.
Still have a question?
Send it over. We answer scoping questions directly, without a discovery call first.
Contact us