CATEGORIES OF USE

CONTACT

“Text” conversational assistant

A text-based conversational assistant is a tool that reads your messages and responds in French (or another language) in writing, much like a human: it summarizes, writes, explains, suggests ideas, helps formulate an email, or prepares an FAQ.

It can be used either:

Online via a web page or app (zero technical requirements),
By API (connected by the IT team to your internal tools) to automate tasks.

Advantages for the "office profile": time savings (writing, summarizing, repetitive responses), available 24/7, consistency.

Points to keep in mind: cost per use (depending on volume), confidentiality (cloud service), and accuracy can be improved on specific topics—you need to know how to proofread.

OpenAI – ChatGPT

To write, summarize, and brainstorm in just a few clicks. Can be used directly in the browser and easily connected to your internal tools if needed.

Google – Gemini

Assistant available on the web and as a mobile app. Practical for answering questions, generating ideas, and integrating into your processes.

Anthropic – Claude

Appreciated for clear answers and long summaries. Works in chat and can be connected to your business solutions.

Perplexity

A response engine that shows its sources. Great for monitoring and searching for quick and reliable information.

xAI – Grok

Assistant linked to X (formerly Twitter). Available as an app and can be connected to your tools for customized uses.

Cohere (Toronto)

Canadian company specializing in language models. Designed to be easily connected to organizational systems.

IBM watsonx (Canada)

Suite of AI tools available in data centers across the country. Useful when data localization and governance are essential.

Mila – Quebec AI Institute (Montréal)

Major AI research hub. Source of talent, collaborative projects, and transfers to industry.

Code generation

They are intelligent assistants that transform your normal sentences into small programs.
You write: "Take a small action that puts this list in order" or "Fix what's not working" → the tool offers a solution ready to try and can explain what's wrong.

We open a website or an application, and we describe what we want in simple language.

OpenAI (assistants) : you explain your need, it suggests a procedure and a solution.
Google – Gemini Code Assist (Colab) : Works in the browser; useful for testing small ideas.
Replit — Replit Code / Agent : a web platform for trying things out and collaborating easily.

When to choose them?
When you want to move quickly, get a first result or understand what is happening without installing anything.

Designed for those who work with code every day.

GitHub Copilot / Microsoft Copilot : Offers ideas while writing and helps improve the result.
Amazon CodeWhisperer : Very useful for teams already working with Amazon tools.
JetBrains AI Assistant : Integrated into software used by many developers.
Sourcegraph Cody : Leverages company code to help better.
Meta Code Llama, StarCoder, DeepSeek Coder : more “technical” options, useful for teams that want a lot of control.

When to choose them?
When a technical team wants to speed up its daily work and integrate with its professional tools.

These assistants speed up work, but do not replace human verification (quality, security, compliance).

Start small : one simple request, one small result, then expand if it really helps you.

Do not include sensitive information if your organization does not allow it.

If you give me a concrete example (e.g. “rename files”, “put a list in order”, “summarize a table”), I will write you a copy-paste ready request to try it out right away.

Image generation

Visual assistants that turn a phrase (“a cat in a suit on a skateboard, retro poster style”) into a ready-to-use image.
Useful for posts, presentations, campaign ideas, product mockups, internal illustrations, etc.

Why is it practical?

Ultra-fast : visuals in seconds.
Economical for simple everyday needs.
Easy to iterate : test several versions and keep the best one.

Keep in mind : Check usage rights, brand guidelines, and avoid sensitive topics. It often takes two or three tries to get exactly what you want.

Open a website/app, describe the image, and off you go. Ideal for marketing, communications, and SMEs.

DALL·E (OpenAI) — “general” generator for on-demand images (posts, slides, concepts). Also available directly from the chat.
Midjourney — very creative for mood boards and quick concepts.
Adobe Firefly — perfect if you use Photoshop / Illustrator; assisted generation and editing.
Canva Magic Studio — ideal for social media and presentations; generation + editing in Canva.
Microsoft Designer — Create banners, posts, illustrations in just a few sentences.
Playground — easy to quickly create and test variations.
Leonardo — useful for product concepts, presentation assets, and various styles.
Ideogram — good for stylized visuals and text in images (posters, ads).
Gemini (Google) – “Nano Banana” — image generation/editing integrated into Gemini, with invisible watermark for traceability.
Dream by WOMBO (Canadian company) — very simple web and mobile app for stylish one-sentence visuals.
Getty Images – Generative AI — designed for secure commercial use (licenses + protections).
Shutterstock – Generative AI — “image bank + AI” with useful legal compensation.

When to choose them?

Quickly produce beautiful visuals without graphic skills.
Decline a message in story / post / banner.
Look for visual ideas for a presentation or campaign.

More control, integration, and “advanced” options. Often with internal support.

Stable Diffusion (Stability AI) — very flexible: online or hosted internally (to control settings and confidentiality).
Flux (Black Forest Labs) — designed for integrations and sustained production (team-oriented).
Imagen (Google) / Emu (Meta) — powerful technologies; access depends on their ecosystem offerings/tools.

When to choose them?

Need for control (privacy, personalization, automation).
Large-scale production : industrialize creation and integrate AI into your in-house tools.

Video generation

Video assistants that turn a sentence (“a drone flies over a forest at sunrise, cinema-style”) into a ready-to-use clip.
Useful for teasers, social media posts, presentations, product mockups, training, storyboards, etc.

Why is it practical?

Very fast : minutes instead of hours/days of production.

Economical for simple needs and prototypes.

Easy to iterate : test several versions and keep the best one.

Keep in mind : Check usage rights, brand guidelines, and avoid sensitive content. The first draft isn't always perfect, so plan on 2–3 tries.

Open a website/app, describe the scene, and off you go. Perfect for marketing, communications, training, and SMEs.

OpenAI — Sora

Rich, realistic videos from a single source. Great for storyboards, short ads, and posts. (API still limited, but the interface is mainstream.)

Google DeepMind — Veo (via Gemini)

Direct generation in the Gemini app/website (Pro/Ultra plans). High-quality clips for presentations, social media, and concepts.

Runway — Gen

Create clips from text, images, or short videos. Ideal for teasers and concepts that need quick validation.

Pika

Ultra-fast creation of short clips (starting text or image). Great for social media.

Luma AI — Dream Machine

Consistent sequences with believable camera movements. Good for concepts, commercials, presentations.

Synthesia

Transforms a script into a video with avatar and voice (multilingual). Perfect for training and internal communication.

Description

Word-like editing: edit the text, and the video adapts. Ideal for tutorials, interviews, filmed podcasts, and subtitles.

Microsoft Designer (simple video)

Express generation and design for animated banners, posts, mini-videos ready to publish.

Kuaishou—Kling

Realistic clips from text or images. Useful for teasers and social content (availability may vary by region).

When to choose them?

Produce short videos quickly without a video team.
Test ideas before heavier production.
Adapt a message into several formats (story, post, video slide).

More control, integration, and advanced options. Often with the support of an internal team.

Stability AI — Stable Video Broadcast

Animates still images into clips. Available as an online service, API, and deployable options (control/privacy).

Google DeepMind — Veo (via API)

Access via Gemini API / Vertex AI to automate (formats, durations, parameters) and integrate into your pipelines.

Meta — Emu Video/Edition

Short video generation/editing from text (and sometimes images). Access is rather selective (R&D, dedicated programs).

When to choose them?

Need for control (privacy, personalization, automation).

Industrialize video creation and integrate AI into your tools/pipelines.

Audio & Music

Tools that transform voice and text :

Voice → Text (transcription) to get reports, subtitles, notes.
Text → Voice (synthetic voice) to create believable voice-overs in multiple languages.
Text → Music/Sounds to generate custom music, jingles and sound effects.

Why is it practical?

Fast : minutes rather than hours.
Economical : perfect for everyday needs and prototypes.
Multilingual : subtitles and voices in several languages.
Accessible : Many tools can be used in the browser, without special software.

Keep in mind : usage rights (voice/music), consent for any voice cloning, respect for the brand charter and sensitive data.

You open a site, upload a file or paste some text, and off you go.

Synthetic voice (Text → Voice)

ElevenLabs TTS – very natural voices, multilingual, tone/emotion adjustments.
Murf – Fast voiceovers for presentations/videos, easy to use.
PlayHT – large selection of voices and languages; advanced options (rhythm, style).
Resemble AI (Resemble Voice) – voice creation/cloning (with consent), useful for brand voice.

Music & Sounds (Text → Music/Sound Effects)

Suno – creates complete songs/jingles from a simple brief.
Udio – generates custom songs (styles, durations).
AIVA – custom instrumental music (videos, games, advertising).
Beatoven – royalty-free music to suit your mood.
Soundful – original tracks for videos/podcasts, easy for professional use.
Stable Audio (Stability AI – web version) – music/effects in a few clicks, with duration options.

When to choose them?

Quickly produce a voiceover, subtitles, or background music.
Test several versions and choose the best one.
Marketing, communications, training, and SME teams.

More control, automation and integration with internal tools.

Transcription

OpenAI Whisper (open source & API) – very strong in multilingual; ideal if you want to integrate transcription into your own tools.

Synthetic voice (cloud APIs)

Google Cloud Text-to-Speech (WaveNet) – large catalog of voices, fine-grained control (API).
Microsoft Azure Neural TTS – natural voices, branded voices possible (API).
OpenAI TTS / GPT-4o Voice – expressive voices, useful for mobile assistants/apps (API).

Music & sounds (tech / specific access)

AudioCraft (Meta – MusicGen/AudioGen, open source) – music/sound effects generation with customization and internal deployment.
Stable Audio (Stability AI – API) – on-demand generation in your apps (automation, batches).
Google Lyria – music generation via Google interfaces; integration/API remains more regulated depending on the offer.

When to choose them?

Need for control (confidentiality, personalization, industrialization).
Automate production (pipelines, batches, internationalization) and integrate AI audio into your products/services.

Translation & Transcription

Tools that :

Translate your texts, documents, websites or messages into other languages.
Transcribe speech to text (meetings, podcasts, calls) and can summarize the essentials.

Why is it useful?

Fast : minutes rather than hours.
Economical : perfect for everyday needs and standby.
Multilingual : 30+ to 100+ languages depending on the tool.
Accessible : Often in the browser or via an app.

Keep in mind : always reread important content (legal, health, sensitive marketing), check confidentiality and usage rights (data, voice).

You open a website/app, upload a file or paste some text, and off you go.

Translation

Google Translate (GNMT) – very versatile (text, documents, sometimes images). Ideal for quick understanding and everyday translation.
DeepL – appreciated for the quality of style (Word/PowerPoint documents retaining formatting, glossaries).

Transcription of meetings / notes

Otter.ai – takes automatic meeting notes, identifies speakers, outputs summaries and keywords.

When to choose them?

Translate emails, web pages, office files.
Get a clean report after a meeting or webinar.
Save time preparing a “first draft” multilingual version.

More control, automation and integration with internal tools.

Transcription (speech → text)

OpenAI Whisper (open source & API) – very strong in multilingual; possible on-premise for more confidentiality.
AssemblyAI ASR – turnkey API service (streaming, diarization, keyword detection).
NVIDIA NeMo ASR – focused on on-premise deployment (data control and security).

Translation (text ↔︎ text / stream)

Amazon Translate – Cloud API, pay-as-you-go, easily integrated into apps/sites.
Microsoft Azure Translator / Speech to Text – real-time translation + transcription with Microsoft cloud guarantees.
Meta Seamless / NLLB (families) – very broad language coverage (including lesser-known languages), open source and via integrations.

When to choose them?

Need for privacy/data residency or automation (tickets, chat, catalog, subtitles).
Industrialize : large volumes, integration with workflows and products (applications, intranet, CRM, call center).