VideoSDK

Real-time AI, voice and video infrastructure for developers.

[VideoSDK @video_sdk]

You don’t like the voice of your realtime model. But you can’t change it. Can’t fix pronunciation. Can’t clean transcripts before LLM. That’s where voice agents fail.

So we rebuilt the pipeline.

Introducing VideoSDK Agents v1 - Prism

Explore Agent v1: https://t.co/hTLy6LBTBp

[VideoSDK @video_sdk]

"Please hold while I transfer you…" 🙄

Then you repeat everything to a stranger.

VideoSDK Warm Transfer kills the cold handoff - the human agent picks up with full context + sentiment already loaded.

Watch it happen 👇

[VideoSDK @video_sdk]

Read how Warm Transfer works in depth:
https://t.co/sgkM9xOy2f

[VideoSDK @video_sdk]

"Please hold while I transfer you…" 🙄

Then you repeat everything to a stranger.

VideoSDK Warm Transfer kills the cold handoff - the human agent picks up with full context + sentiment already loaded.

Watch it happen 👇

[Caoimhe @caoimhe_anam]

@video_sdk Thanks to @SagarKava_ @Arjun_Kava and the VideoSDK team on the collaboration.

Check it out: https://t.co/1BwV4Xg1FR

[Caoimhe @caoimhe_anam]

Anam interactive avatars are now officially on @Video_SDK.

Our integration brings real-time avatars into VideoSDK agent pipelines.

Video > voice > text: 70% user preference over voice-only across customer deployments.

[VideoSDK @video_sdk]

Docs shouldn’t be searched.
They should answer.

So we built an MCP Server for VideoSDK Agents Doc🚀

Ask anything. Get exactly what you need.
Docs that respond > docs you search.

🔗Explore - https://t.co/6xiQL8HUL2

[Arjun Kava @Arjun_Kava]

You don’t like the voice of your realtime model. But you can’t change it.

Can’t fix pronunciation. Can’t clean transcripts before LLM. That’s where voice agents fail.

So we rebuilt the pipeline.

[VideoSDK @video_sdk]

Which means the same system can be used to build:

• transcription pipelines
• voice copilots
• chat + voice agents
• fully autonomous voice agents
• realtime agents

Read more about Agent v1 here : https://t.co/hTLy6LBTBp

[VideoSDK @video_sdk]

You don’t like the voice of your realtime model. But you can’t change it. Can’t fix pronunciation. Can’t clean transcripts before LLM. That’s where voice agents fail.

So we rebuilt the pipeline.

Introducing VideoSDK Agents v1 - Prism

Explore Agent v1: https://t.co/hTLy6LBTBp

[VideoSDK @video_sdk]

• Unified Pipeline
• pipeline hooks - control any stage (audio, text, turns)
• Hybrid Pipeline - Mix any stack - STT, LLM, TTS, knowledge base
• Compose anything - STT , LLM-only, full voice, realtime
• Built-in observability - metrics, logs, traces
• Avatar support plug

[VideoSDK @video_sdk]

🚀 𝗚𝗲𝗺𝗶𝗻𝗶 𝟯.𝟭 𝗙𝗹𝗮𝘀𝗵 𝗟𝗶𝘃𝗲 is now supported on VideoSDK AI Voice Agents!

@Google just launched their most capable real-time voice model yet and you can start building with it on VideoSDK today.

Check out the Docs now: https://t.co/v1LGiaHFkr

[VideoSDK @video_sdk]

We’re excited to introduce @Anam__ai integration with VideoSDK AI Voice Agents 🚀

You can now add real-time, expressive AI avatars to your voice agents  with natural lip sync and sub-second responses.

👉 𝗘𝘅𝗽𝗹𝗼𝗿𝗲 𝘁𝗵𝗶𝘀 𝗶𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗶𝗼𝗻 : https://t.co/mTWk6ELHNq

[VideoSDK @video_sdk]

𝗪𝗵𝘆 𝘁𝗵𝗶𝘀 𝗺𝗮𝘁𝘁𝗲𝗿𝘀:

- Higher engagement with video-first experiences
- Natural, real-time conversations with low latency
- Best-in-class realism powered by Anam’s CARA model
- bring your own LLM, customize personas, clone voices and support 50+ languages

[VideoSDK @video_sdk]

We’re excited to introduce @Anam__ai integration with VideoSDK AI Voice Agents 🚀

You can now add real-time, expressive AI avatars to your voice agents  with natural lip sync and sub-second responses.

👉 𝗘𝘅𝗽𝗹𝗼𝗿𝗲 𝘁𝗵𝗶𝘀 𝗶𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗶𝗼𝗻 : https://t.co/mTWk6ELHNq

[VideoSDK @video_sdk]

Whether you're building customer support bots, AI meeting assistants, or multilingual voice apps - @video_sdk AI Voice Agents gives you the fastest path from idea to a live voice experience. And now it runs on the most powerful real-time model @Google has ever shipped.

[VideoSDK @video_sdk]

🚀 𝗚𝗲𝗺𝗶𝗻𝗶 𝟯.𝟭 𝗙𝗹𝗮𝘀𝗵 𝗟𝗶𝘃𝗲 is now supported on VideoSDK AI Voice Agents!

@Google just launched their most capable real-time voice model yet and you can start building with it on VideoSDK today.

Check out the Docs now: https://t.co/v1LGiaHFkr

[VideoSDK @video_sdk]

𝗚𝗲𝗺𝗶𝗻𝗶 𝟯.𝟭 𝗙𝗹𝗮𝘀𝗵 𝗟𝗶𝘃𝗲 𝗯𝗿𝗶𝗻𝗴𝘀:

- Ultra-low latency audio-to-audio responses
- Acoustic nuance detection (pitch, pace, tone)
- 70+ languages supported in real time
- Improved tool calling
- 2x longer conversation memory

[Anam.ai @Anam__ai]

Anam interactive avatars are now natively supported on @Video_SDK.

Add real-time video avatars to VideoSDK agent pipelines. Under 10 lines of Python. Sub-second response times. Works with your existing STT, LLM, and TTS.

70% user preference over voice-only. CARA model leads all tested providers on visual quality, lip sync, and overall experience (https://t.co/2Oj82zaSXZ).

https://t.co/H6S3h2r8vV

@SagarKava_ @Arjun_Kava

[Arjun Kava @Arjun_Kava]

1/ Introducing AI Voice Agent Cloud Deployments via CLI + Dashboard ⚡️

Deploying voice agents shouldn’t feel complicated.

[VideoSDK @video_sdk]

- No multiple provider accounts to manage
- Works across telephony, web, mobile, and IoT
- Built for true real-time, low-latency conversations
- Supported models - gemini, sarvam AI, Deepgram, Cartesia

[VideoSDK @video_sdk]

Announcing VideoSDK Inference: One Magic API for Every Voice AI Model 🎉

Maintaining multiple accounts for speech recognition, language models, and speech synthesis, each with its own keys, quotas, billing, and APIs

👉 Explore VideoSDK Inference : https://t.co/8ZoDliyyCe