Carter Huffman

Here is the cheapest transcription API in the world. Period. Let's be honest: Transcription APIs are among the most overpriced components of voice AI infrastructure right now. If you are building a voice agent or any sort of conversational AI, a huge chunk of your bill is just transcription. Get more than a few users, and the cost no longer makes sense. The Modulate team launched Velma Transcribe, and I think it's simply the cheapest API I've seen: $0.13 per 1,000 minutes of audio transcription. This is just ridiculous pricing, and every other API I've ever used is way more expensive than that. If you are building voice applications, this is gold.

Most AI builders underestimate how expensive transcription becomes once their product starts scaling. For a lot of teams, speech-to-text quietly turns into one of the biggest infrastructure bills. Just tested **Velma Transcribe** from Modulate and the pricing is surprisingly low compared to typical STT APIs. If you're building: • voice agents • AI assistants • support bots transcription costs matter a lot. Velma claims dramatically lower pricing while still competing with models like Deepgram, ElevenLabs, and AssemblyAI. You can compare the models side by side here: speechtxt.com/?utm_source=x&…

I can’t believe nobody caught this. Speech-to-text pricing just quietly got obliterated. A new model from @modulate_ai is charging: $0.03 per hour of audio. For context, that’s: • 5× cheaper than @AssemblyAI • 8.6× cheaper than @DeepgramAI • 13.3× cheaper than @ElevenLabs If you’re building: • voice agents • meeting assistants • call center analytics • conversation AI This isn’t a small optimization. It changes the entire economics of voice products. Worth testing it yourself using this comparison tool: speechtxt.com/?utm_source=x&… Velma Transcribe API by Modulate: modulate.ai/lp/transcripti…

🚨 Transcription might be getting a lot cheaper. A new speech-to-text API called Velma Transcribe just launched from @modulate_ai . For many AI companies, transcription is one of the biggest recurring infrastructure costs. This could change that. 🧵

Voice AI might be entering its “AWS moment” with this startup. For years, one part of the voice stack has stayed weirdly expensive: transcription infrastructure. Every voice agent, AI assistant, meeting tool, and call-center automation depends on speech-to-text. Now @modulate_ai just launched Velma Transcribe, a new transcription API built for real conversational audio. And the big claim: It’s 10–90x cheaper than typical speech-to-text APIs while maintaining strong real-world accuracy. That matters because transcription is often one of the largest recurring costs in voice AI systems. If that cost collapses, the ripple effects could be huge: • cheaper voice agents • more conversational AI apps • better meeting intelligence tools • scalable call-center analytics • faster experimentation for builders Velma is designed to compete with models like @DeepgramAI Nova-3, @elevenlabs Scribe v2, and @AssemblyAI Universal-2 on both cost and accuracy. It also includes production-ready features like: • streaming transcription • batch processing • speaker diarization • timestamps • emotion & accent detection • 70+ language support If transcription infrastructure actually becomes 10x cheaper, we might see an explosion of new voice products. Explore it here: modulate.ai/lp/transcripti… Compare STT models here: speechtxt.com/?utm_source=x&… Curious to see how this reshapes the voice AI stack.

Voice infrastructure costs are about to get interesting. For a long time the default transcription stack for a lot of AI apps has been things like Deepgram, ElevenLabs, or AssemblyAI. But I just tested Modulate’s new **Velma-2 transcription** model and the pricing surprised me. It’s running at about **$0.13 per 1,000 minutes**. That’s dramatically cheaper than most STT APIs I’ve seen. You can compare the models side by side here: speechtxt.com/?utm_source=x&…

If you're building anything with voice right now, this is worth trying. @modulate_ai just released this API that's much cheaper than most options: 🦾 taaft.co/modulate They also built a tool where you can compare it against platforms like: Deepgram ElevenLabs AssemblyAI

🚨BREAKING: A startup called Modulate just nuked one of the biggest hidden costs in AI voice products. For years companies like Deepgram, AssemblyAI, and ElevenLabs Scribe have been charging premium prices for transcription. Now there’s an API that’s 10–90x cheaper. Not slightly cheaper. Orders of magnitude cheaper. This could change voice AI economics overnight. Here's everything you need to know 👇

Speech transcription just got much cheaper. The reason is architectural. @moduate_ai Velma uses Dynamic Ensemble Blocks - multiple specialized models that dynamically switch during inference instead of relying on one giant model. Result: 14.9% WER (AMI) at $0.13 / 1k minutes. We launched the Velma Transcribe API today: bit.ly/3Nmh1WK Test Velma-2 against leading models here: bit.ly/3P3SxCj

I had so much fun on the Build AI podcast talking about @modulate_ai 's new Ensemble Listening Model architecture - how to do VoiceAI Better, Faster, and 100x Cheaper than foundational models like ChatGPT or Gemini! Check it out: youtube.com/watch?v=MAbMG7… open.spotify.com/episode/7jWNq8… podcasts.apple.com/us/podcast/gta…

@jasonhiner @Techmeme Better, Faster, and Cheaper #VoiceAI compared to foundational models like #ChatGPT and #Gemini3 - incredible work by our team of engineers and researchers!

Exclusive: Boston startup Modulate challenges frontier labs with a cheaper, more accurate ensemble orchestration technique that could reshape AI (tip @techmeme) thedeepview.com/articles/break…

ChatGPT for voice? It's possible -- but is it what you really want? Modulate is a leading AI tech company in voice analytics, and we do more for understanding real human speech than any LLM can do today. Stay tuned for an exciting announcement next week 👀