Back

Shyam Vijay

Content strategist f... • 4h

Hello folks, I’m Shyam Vijay from MirrorFly (https://www.mirrorfly.com/video-call-solution.php). We build APIs and SDKs that let developers drop video, voice, and chat into their apps. Today we’re launching some big upgrades to our Video Call SDK: Real-time speech-to-text (https://www.mirrorfly.com/conversational-ai/speech-to-text/), AI chatbot + AI voice agent support (https://www.mirrorfly.com/conversational-ai/chatbot/ and Video moderation Quick demo (captions + moderation inside a call): https://www.mirrorfly.com/request-demo.php Why we built this Over the past year, the same requests kept coming up from our customers in telehealth, edtech, and social platforms: 1.“We need live captions and transcripts.” (Accessibility, compliance, note-taking) 2.“We need to automatically catch unsafe or harmful video.” 3.“We want AI-powered assistants inside calls.” (Chatbots and voice agents that can answer questions or guide users in real time) On the surface, it’s just “add transcription” or “add moderation.” But once you’re running at production scale, things get painful fast: Speech-to-text has to be accurate and near-instant. Delayed captions make the feature useless. Handling noisy environments, different accents, and overlapping speakers pushed us into building streaming pipelines optimized down to milliseconds. AI moderation has to run frame by frame, in real time, without tanking call performance. Scaling that across thousands of concurrent calls was way harder than we expected. Chatbots and voice agents need low-latency access to the transcript + call stream, or else they feel robotic. We had to rethink our data flow so LLMs could “listen in” and respond naturally. What’s new in the SDK This update gives developers: Real-time captions with transcript storage/search. Customizable AI moderation (choose which categories to flag/block). AI chatbot integration for text-based agents in calls. AI voice agents that can participate in live conversations. Cross-platform SDKs (Web, iOS, Android). Privacy-first design with HIPAA and GDPR compliance. Try it out You can try it here: https://www.mirrorfly.com/contact-sales.php. We’ve got free developer credits to experiment. Pricing is usage-based (per minute of transcription, moderation, or AI agent time), with volume discounts at scale. All data belongs to our customers. We support 0-day retention and never train on customer data. For healthcare and EU customers, we’re HIPAA and GDPR compliant out of the box.

Reply

More like this

Recommendations from Medial

aavyamedia

Hey I am on Medial • 5m

5 Free AI Tools for Social Media Posts Canva – AI design & captions Adobe Express – Firefly AI posts Simplified – 1-click posts & captions VistaCreate – Free templates Writesonic – AI text + visuals #ai #MarketingTrends #ArtificialInteligence

Reply
1
4
Image Description
Image Description

Aj

Hey I am on Medial • 6m

Anyone interested in building AI voice agents?

2 Replies
3
Image Description

Krish Vaidya

Dreaming big so peop... • 7m

Building and selling agents such as Chatbots and Ai voice calling agents to business owners for a monthly fee also marketing their business with Ai. So ai voice calling agents work like customer support staff what I noticed that businesses spend a lo

See More
1 Reply
4
Image Description
Image Description

Krish Vaidya

Dreaming big so peop... • 7m

So my startup idea is ai agents Building and selling agents such as Chatbots and Ai voice calling agents to business owners for a monthly fee also marketing their business with Ai. So ai voice calling agents work like customer support staff what I n

See More
8 Replies
2
12

Mitsu

extraordinary is jus... • 4m

- Text to mindmaps => My map Ai - Text to infographics => Infography .in - Text to Presentation => Decktopus - Text to YouTube video => Syllaby - Text to image => Reve AI - Text to Ads => Creatify Ai - Text to ideas => Virality AI - Text to Music =>

See More
Reply
1
10
Image Description
Image Description

Havish Gupta

Figuring Out • 1y

OpenAi be like Text to Image: 2021 Text to Text: 2022 Text to Video: 2024 Text to App: 2025 Text to Ai Model: 2026 Text to Physical Object: 2027 Text to Human: 2028 . . . Text to Planet: 2100

9 Replies
33
Image Description
Image Description

MK

Hey I am on Medial • 8m

I think Medial should come up video call and voice call feature as well , because it would be easy to discuss idea with founders rather then exchanging number .

8 Replies
2

Dr Sarun George Sunny

The Way I See It • 11m

Sarvam AI: Empowering Indian Languages with AI Sarvam AI, co-founded by Vivek Raghavan and Pratyush Kumar, focuses on building AI that caters to India’s linguistic diversity. Their Indic Voice Large Language Models (LLMs) allocate 40% of their token

See More
Reply
4

Allison william

Hey I am on Medial • 1m

🗣️ What is an AI Voice Agent? AI voice agents are intelligent systems that can understand, interpret, and respond to human speech—powering everything from customer support bots to voice-enabled apps like Alexa and Siri. 🤖✨ These tools are revoluti

See More
Reply
4
Image Description

Divyraj

Future founder • 9m

is any AI makes text into video?

1 Reply
1

Download the medial app to read full posts, comements and news.