Average Ratings 0 Ratings
Average Ratings 0 Ratings
Description
Grok Voice 3.0 is an advanced speech-to-speech model that is accessible via the Grok Voice Agent API, designed for developers creating real-time voice applications. This model processes both audio and text inputs, delivering spoken audio or text outputs through a WebSocket connection while enabling features such as conversational turn-taking, interruptions, and function execution. Developers have the flexibility to link this model with custom functions and various server-side tools, which may include web searches, X searches, collections, or remote MCP servers. The API is crafted for use in interactive voice agents that enhance customer support and service call experiences, and it is also implemented for voice communication within Tesla vehicles and Grok Bot. Billing operates on a usage-based model, charging for audio by the minute and text input events separately. Additionally, the API documentation provides comprehensive setup instructions along with examples for clients using Python and JavaScript, ensuring that developers can easily integrate the functionality into their applications. This robust feature set empowers developers to create more engaging and efficient voice interaction systems.
Description
Voxtral Small Transcribe 3 is a versatile multilingual speech-to-text model tailored for converting recorded audio into text. Offered via an audio transcription API by Mistral, it incorporates features like timestamp documentation, speaker diarization, and context biasing, enabling developers to input key phrases that enhance the recognition of specific terms and names relevant to particular domains. This model is capable of automatically identifying supported languages or can utilize a specified language code, accommodating over 50 languages in total. It is particularly suitable for application developers focused on creating workflows for transcribing recorded audio, such as meetings, interviews, call transcriptions, customer service systems, and voice-agent processes. Additionally, Mistral outlines options for self-deployment of the Voxtral 3 series, which includes hosting on private clouds or on-premises solutions, making it adaptable to various organizational needs. Furthermore, the flexibility of the model ensures that it remains a valuable tool for enhancing communication efficiency across different sectors.
API Access
Has API
Yes
API Access
Has API
Yes
Integrations
Grok
Yes
Grok Bot
Yes
Grok Voice Agent
Yes
Grok Voice Agent Builder
Yes
Mistral AI
No
Model Context Protocol (MCP)
Yes
Twilio
Yes
Integrations
Grok
No
Grok Bot
No
Grok Voice Agent
No
Grok Voice Agent Builder
No
Mistral AI
Yes
Model Context Protocol (MCP)
No
Twilio
No
Pricing Details
$0.08/min
Audio is $0.08 per minute. Text-only input events are $0.004 each. Server-VAD sessions are billed for the full session duration; push-to-talk sessions are billed for audio sent and received.
Free Trial
Yes
Free Version
No
Pricing Details
$0.004/min
API usage is $0.004 per minute of audio.
Free Trial
No
Free Version
No
Deployment
Web-Based
Yes
On-Premises
No
iPhone App
No
iPad App
No
Android App
No
Windows
No
Mac
No
Linux
No
Chromebook
No
Deployment
Web-Based
Yes
On-Premises
Yes
iPhone App
No
iPad App
No
Android App
No
Windows
No
Mac
No
Linux
No
Chromebook
No
Customer Support
Business Hours
No
Live Rep (24/7)
No
Online Support
Yes
Customer Support
Business Hours
No
Live Rep (24/7)
No
Online Support
Yes
Types of Training
Training Docs
Yes
Webinars
No
Live Training (Online)
No
In Person
No
Types of Training
Training Docs
Yes
Webinars
No
Live Training (Online)
No
In Person
No
Vendor Details
Company Name
SpaceXSI
Founded
2023
Country
United States
Website
x.ai/api/voice
Vendor Details
Company Name
Mistral AI
Founded
2023
Country
France
Website
docs.mistral.ai/models/voxtral-small-transcribe-3-0