Help Center

How can we help you?

Guides, walkthroughs, and answers for every Nuvio product — all in one place.

Getting Started

Build your first AI chatbot in minutes
No code required. Converso lets you create a fully trained AI chatbot by uploading your documents or adding URLs — no developer needed.

Visit converso.nuvione.com and sign up with your work email. No credit card is required to get started — you can build and test your first bot on the free plan.

From your dashboard, click New Bot. Give it a name (e.g., "HR Assistant"), choose a persona tone (Professional, Friendly, or Formal), and select your primary language. You can change all of these later.

Upload PDFs, Word docs, or plain text files — or paste URLs for Converso to crawl. Your bot will be trained on this content. Supported formats: PDF, DOCX, TXT, CSV, and website URLs. You can add more content at any time from the Sources tab.

Once training completes (usually under 2 minutes), use the Preview panel on the right to chat with your bot. Test with questions your users would actually ask. If answers are missing or incorrect, return to Sources and add more relevant content.

Go to Deploy tab. Copy the embed snippet and paste it into your website's HTML (before the closing </body> tag). Or share the hosted link directly with your team. The bot is live instantly — no server restarts needed.

Training & Content

Managing what your bot knows

Converso supports PDF, DOCX, TXT, CSV, and direct website URLs. For best results, ensure documents are text-based (not scanned images). Scanned PDFs can be processed but may have reduced accuracy — use Nuvio Extract first to convert them to clean text if needed.

Go to Sources in your bot settings. Add new files or URLs, remove outdated ones, then click Retrain. Retraining typically completes in 1–3 minutes depending on content volume. Your bot continues to answer questions from its previous training while the new version processes.

This usually means the source content doesn't contain enough detail on that topic. Add a document or FAQ that specifically covers the question, then retrain. You can also use Bot Instructions (in Settings) to guide tone, scope, and fallback behavior — for example, telling the bot to say "Please contact HR directly" when a topic is out of scope.

Yes. Converso detects the language of each user's message and responds in the same language — even if your source documents are in English only. You can also set a forced language in Settings → Language if you want to restrict responses to a specific language.

Embedding & Deployment

Put your bot on your website or share via link

In your bot's Deploy tab, copy the JavaScript snippet. Paste it just before the closing </body> tag on any page where you want the chat widget to appear. The widget loads asynchronously and won't affect your page speed. Works with any CMS — WordPress, Webflow, Squarespace, and custom HTML.

Yes. Under Settings → Appearance, you can set your brand color, upload a bot avatar, customize the greeting message, and change the widget position (bottom-right or bottom-left). On Pro and Business plans, you can also white-label the widget — removing the "Powered by Converso" badge.

Yes. Converso provides a REST API for sending messages and receiving responses programmatically — useful for integrating into Slack, Microsoft Teams, CRMs, or custom applications. Find your API key and full documentation under Settings → API Access. API access is available on Business and Enterprise plans.

Go to Analytics in your bot dashboard. You'll see total conversations, unique users, most common questions, unanswered queries (great for finding content gaps), and satisfaction ratings if you've enabled the feedback prompt. You can export conversation logs as CSV for deeper analysis.

Troubleshooting

Common issues and how to fix them

Check that: (1) the embed snippet is placed before the closing </body> tag, (2) your bot is set to Published (not Draft) in the Deploy tab, and (3) your website's Content Security Policy (CSP) isn't blocking external scripts. If using a CMS plugin, ensure JavaScript is enabled for the page. Still not working? Submit a support ticket with your domain and bot ID.

The content covering that topic may be missing, too brief, or in a format the bot couldn't extract (e.g. scanned images). Add a specific document or FAQ entry that explicitly answers the question, retrain, and test again. You can also view Unanswered Questions in Analytics to see the full list of gaps.

Message limits reset at the start of each billing cycle. To increase your limit immediately, upgrade your plan under Settings → Billing. If you're on the free plan, upgrading to Starter or Pro will significantly increase your monthly message allowance. Contact us if you need a custom volume plan.

Getting Started

Extract structured data from your first document
What Extract does: Upload any document — invoice, contract, form, report — and Extract uses AI to pull out structured fields automatically. No templates required.

Visit extract.nuvione.com and sign in with your Nuvio account. New users can sign up for free — no credit card required for the starter tier.

Click New Extraction and drag-and-drop your file or click to browse. Supported formats: PDF, PNG, JPG, TIFF, DOCX. For best results, upload high-resolution files. Multiple files can be uploaded at once for batch processing.

Select a pre-built schema (Invoice, Contract, Receipt, Form, ID Document) or define your own custom fields. Custom schemas let you specify exactly which fields to extract and in what format. Once a schema is saved, reuse it across all future extractions of the same document type.

Click Run Extraction. Results appear in seconds for single files, minutes for large batches. Review extracted fields side-by-side with the source document. Download results as CSV, JSON, or Excel — or push them directly to your connected systems via API or webhook.

Supported Document Types

What Extract can process

Extract automatically identifies: vendor name, invoice number, date, due date, line items, quantities, unit prices, subtotals, tax, and total amount due. Works across different invoice layouts without needing separate templates per vendor.

For contracts, Extract pulls party names, effective dates, termination dates, key obligations, payment terms, and clause summaries. Use a custom schema to flag specific clauses (e.g., limitation of liability, renewal terms) across a library of contracts in one batch run.

Extract includes OCR (optical character recognition) for scanned PDFs and images. Handwritten documents are supported with reduced accuracy. For best OCR results: use 300 DPI or higher resolution, ensure good contrast, and avoid skewed or rotated pages. Accuracy confidence scores are shown per field so you can spot and correct low-confidence extractions.

Upload hundreds of documents at once using the Batch tab or via API. All documents are processed in parallel. Results are collected into a single downloadable file (CSV or JSON). Batch processing is available on Growth and Business plans. API batch limits depend on your plan tier.

API & Integrations

Connect Extract to your existing systems

Go to Settings → API Access in your Extract dashboard. Click Generate API Key. Your key is shown once — copy and store it securely. Pass the key as a Bearer token in the Authorization header of each API request. Full API documentation is linked from the same page.

Configure a webhook URL under Settings → Webhooks. When an extraction completes, Extract sends a POST request to your URL with the results as JSON. This lets you trigger downstream actions automatically — updating a CRM, sending a Slack message, or inserting records into a database — without polling the API.

Yes — use the webhook trigger in Zapier or Make to receive extraction results and pass them to Google Sheets, Salesforce, HubSpot, Airtable, or any other connected app. If you need a native Zapier app or pre-built integration template, contact our team and we'll help you set it up.

Troubleshooting

Common extraction issues

Check the confidence score next to each field — fields below 80% may need review. Common causes: low-resolution scan, unusual document layout, or handwritten content. Try improving scan quality, or use a custom schema to help the AI identify specific fields by name. You can manually correct results before downloading.

Maximum file size is 50MB per file. If your file exceeds this, compress the PDF or split it into smaller chunks. Unsupported formats (e.g. Excel, PowerPoint) should be converted to PDF first. Password-protected PDFs must be unlocked before uploading.

Getting Started

Process your first image batch in minutes
What Imago does: Upload hundreds or thousands of images and apply transformations — resize, convert, watermark, AI-tag, classify — all in a single automated batch job.

Visit imago.nuvione.com and sign in with your Nuvio account. The free tier allows up to 100 images per batch — upgrade for higher volumes.

Click New Batch and upload individual images, a folder, or a ZIP archive. Supported formats: JPG, PNG, WEBP, TIFF, BMP, GIF. You can also connect a cloud storage folder (Google Drive, S3, Dropbox) as a source on Business plans.

Choose which operations to apply: Resize (set dimensions or percentage), Convert (change format), Compress (reduce file size), Watermark (add text or image overlay), AI Tagging, or Rename (using dynamic naming patterns). Operations are applied in sequence — drag to reorder them.

Click Run Batch. Progress is shown in real time. When complete, download all processed images as a ZIP, or export to your connected cloud storage. A processing report shows success/failure per image, file size changes, and any AI tags applied.

Processing Options

Resize, convert, watermark, and more

Set a target width, height, or both. Choose a fit mode: Contain (letterbox to fit within dimensions, preserving ratio), Cover (crop to fill dimensions exactly), Stretch (distort to exact size), or Percentage (scale by a factor, e.g. 50%). You can also set a maximum file size and Imago will automatically find the right dimensions to meet it.

Add a text watermark (with custom font, size, color, and opacity) or an image watermark (upload your logo). Set position to any corner, center, or tiled across the entire image. Watermarks are applied non-destructively to the output file — your originals are never changed.

Use the Rename operation to apply consistent naming conventions across your batch. Available tokens: {original} (original filename), {index} (sequential number), {date}, {width}, {height}, and any custom AI tag values. Example pattern: product_{index}_{width}x{height}product_001_800x600.jpg

AI Features

Automatic tagging, classification, and analysis

Enable AI Tagging in your pipeline. Imago's vision model analyzes each image and generates descriptive tags — objects, scenes, colors, moods, and product categories. Tags are embedded in the image metadata (EXIF/IPTC) and exported in the batch report CSV. Use these tags to power search, categorization, and DAM (Digital Asset Management) workflows.

Yes. Use the AI Classification operation and define your own category labels (e.g. "outdoor", "indoor", "people", "product"). Imago assigns each image to the best-matching category and organizes output files into subfolders accordingly. Custom classifiers trained on your own image library are available on Enterprise plans.

Troubleshooting

Common batch processing issues

After a batch completes, click View Report to see the per-image error log. Common causes: corrupted source file, unsupported color mode (e.g., CMYK — convert to RGB first), or file exceeding 100MB. Failed images are listed separately and can be re-processed individually after fixing the source.

Processing time scales with batch size, image resolution, and the number of operations. AI Tagging and Classification add processing time per image. Very large batches (1000+ images) on high-res source files can take 10–30 minutes. You'll receive an email notification when the batch is complete so you don't need to stay on the page.

Getting Started

Up and running in minutes
Before you begin: Nuvio Prep requires an API key from at least one AI provider — Claude, Gemini, or OpenAI. Your key is stored locally and never sent to Nuvio's servers.

Head to the Nuvio Prep download page and register with your email. No credit card is required for Nuvio itself — you only pay the AI provider you choose to use, directly at cost.

See the Connecting AI Providers section below for step-by-step instructions for Claude, Gemini, and OpenAI. Each takes just a few minutes and providers offer either a free tier or pay-as-you-go billing.

Open Nuvio Prep, go to Settings → AI Provider, paste your key, select the model, and click Save & Verify. Nuvio will confirm the connection is working before you proceed.

Go to Practice → New Interview, paste the job description, choose a question track (Technical, Behavioral, Leadership, or Scenario), and press Start. The AI will ask questions, listen to your answers, and deliver a detailed score report at the end.

Claude excels at behavioral and cultural-fit interviews — conversational, nuanced, and gives detailed coaching feedback. Gemini offers a generous free tier, ideal for starting at no cost. OpenAI GPT-4o is strongest for technical and coding interviews. You can switch providers at any time without losing your history.

Connecting AI Providers

Get your API key from Claude, Gemini, or OpenAI
Security reminder: Treat your API key like a password — never share it publicly. If you suspect it's exposed, rotate it immediately from the provider's dashboard and update it in Nuvio settings.
C
Claude
by Anthropic

Best for behavioral & cultural-fit interviews. Delivers nuanced, thoughtful coaching and detailed feedback.

How to get your key
1
Visit console.anthropic.com and create a free account.
2
In the sidebar, go to API Keys → Create Key.
3
Name it (e.g. "Nuvio"), copy immediately and store safely.
4
Add a payment method under Billing.
G
Gemini
by Google

Generous free tier — great for getting started at zero cost. Strong reasoning with a familiar Google sign-in.

How to get your key
1
Go to aistudio.google.com and sign in.
2
Click Get API Key → Create API Key.
3
Select or create a Google Cloud project.
4
Copy the key and paste it into Nuvio Prep.
O
OpenAI
by OpenAI

Best for technical & coding interviews. GPT-4o is the industry benchmark for software engineering roles.

How to get your key
1
Sign up at platform.openai.com.
2
Avatar → API Keys → Create New Secret Key.
3
Copy immediately — shown only once.
4
Add credits under Billing → Add to credit balance.

Features

Making the most of Nuvio Prep

After each answer, the AI evaluates clarity, relevance, structure (STAR method), and confidence level. A summary scorecard is shown at the end of every session, broken down by competency area.

Yes — paste the job description when starting a new session. Nuvio Prep tailors questions to the role's required skills and known interview formats. Choose a track: Technical, Behavioral, Leadership, or Scenario.

Many employers use AI voice screening tools (HireVue, Paradox, etc.) that transcribe and score your answers automatically. Nuvio Prep's Speech Coach shows your transcription accuracy and highlights patterns to fix before the real interview.

When you paste a job description and resume, Nuvio Prep automatically creates a local folder named after the company and role — saving your tailored resume, cover letter, the JD, and generated questions. All accessible even months later from Progress → Interview History.

Troubleshooting

Common issues and how to fix them

Double-check that you copied the key without extra spaces. If the error persists, generate a new key in the provider's dashboard and update it in Settings → AI Provider. Also confirm you have an active billing method on the provider's side.

Usually a rate-limit or quota issue on the provider's side, especially on free-tier accounts. Try upgrading your provider plan, switching to a faster model, or trying a different provider.

Completely private. Nuvio Prep is a desktop application — your resume, audio recordings, transcriptions, job folders, and generated documents are stored locally on your machine only. We never receive or store your personal data.

Nuvio Prep does not manage your AI provider billing — charges come directly from Anthropic, Google, or OpenAI. Check your provider's usage dashboard. Setting a monthly spend limit in the provider portal is strongly recommended.

No results found

Try different keywords, or browse the sections in the sidebar.
Can't find what you need? Our support team can help.

Submit a Support Ticket