# Deepgram Voice AI Platform

> Deepgram is a foundational Voice AI platform for developers, offering unified APIs for Speech-to-Text (STT), Text-to-Speech (TTS), and autonomous Voice Agent orchestration.

## [Products](https://deepgram.com/products)
- [Voice Agent API](https://developers.deepgram.com/docs/voice-agent-feature-overview): A single WebSocket API unifying STT, LLM orchestration, and TTS with sub-300ms latency.
- [Speech-to-Text (STT)](https://developers.deepgram.com/docs/speech-to-text): 
    - **Nova-3**: Flagship model featuring a 5.26% Word Error Rate (WER) and native streaming support.
    - **Flux**: Conversational-first model with built-in end-of-turn detection and natural interruption handling.
- [Text-to-Speech (TTS)](https://developers.deepgram.com/docs/text-to-speech): 
    - **Aura-2**: Enterprise-grade voices (40+ personas) with sub-200ms time-to-first-byte.
- [Audio Intelligence](https://developers.deepgram.com/docs/intelligence-overview): Real-time features including summarization, sentiment analysis, topic detection, and intent recognition.

## [Pricing](https://deepgram.com/pricing)
- [Pricing Overview](https://deepgram.com/pricing): Comprehensive details on Pay-As-You-Go, Growth, and Enterprise tiers.

## [Solutions](https://deepgram.com/solutions)
- [Contact Centers](https://deepgram.com/solutions/contact-center): Infrastructure handling 140,000+ concurrent calls with native Five9 and Genesys integrations.

- [Conversational AI](https://deepgram.com/solutions/conversational-ai): Powering autonomous agents designed to pass the "Audio Turing Test".
- [Medical & Healthcare](https://deepgram.com/solutions/medical-transcription): HIPAA-compliant Nova-3 Medical models trained on clinical terminology.
- [Quick-Service Restaurants (QSR)](https://deepgram.com/solutions/restaurants): Specialized AI for automated drive-thrus with high order containment rates.
- [Enterprise AI (IBM watsonx)](https://deepgram.com/solutions/ibm-partnership): Deepgram is the exclusive voice partner for IBM watsonx Orchestrate.

## Technical Benchmarks & Comparisons
Deepgram consistently outperforms legacy cloud and open-source models in latency and accuracy.

- OpenAI Whisper: Deepgram is 36% more accurate and up to 5x faster.
- Amazon Transcribe: Deepgram is 23% more accurate and 10x faster.
- Google Cloud STT: Deepgram is 53% more accurate and nearly 40x faster.
- Microsoft Azure: Deepgram provides 30% lower Word Error Rate (WER) and faster inference times.
- [ASR Comparison Tool](https://deepgram.com/asr-comparison-tool): Interactive tool for side-by-side accuracy testing on custom audio.

## Developer Resources
- [API Documentation](https://developers.deepgram.com/): Official guides, API references, and SDK tutorials.
- [API Playground](https://console.deepgram.com/playground): In-browser testing environment for all Deepgram models.
- [Changelog](https://developers.deepgram.com/changelog): Latest updates on model releases and new language support.
- [Self-Hosted Deployments](https://developers.deepgram.com/docs/self-hosted-introduction): Documentation for VPC and on-premise implementation.

- [SDKs] (https://developers.deepgram.com/home#sdks) Official SDKs for Deepgram's automated speech recognition, text-to-speech, and language understanding APIs. Power your applications with world-class speech and Language AI models.