VibeVoice 7B API

Q: What is the VibeVoice 7B model ID?

Use vibevoice/7b as the model parameter. Requests go to https://api.aimlapi.com/v1/tts with your AI/ML API key.

VibeVoice 7B is a high-quality text-to-speech model that supports multi-speaker dialogue generation with natural-sounding voices.

Modalities: TextAudio
Price: $0.0007per second
Model type: Text to speech
Released: Jan 5, 2026

VibeVoice 7B overview

VibeVoice 7B by Microsoft

VibeVoice 7B API enables natural voice generation with an advanced neural architecture for voice-driven applications.

VibeVoice 7B specs

Model ID: vibevoice/7b
Developer: Microsoft
Input: Text
Output: Audio
Endpoint: POST https://api.aimlapi.com/v1/tts

VibeVoice 7B is a cutting-edge AI-powered voice synthesis model engineered for generating highly natural, expressive, and context-aware speech outputs. Tailored for developers, content creators, and enterprises, VibeVoice 7B delivers versatile voice solutions across industries including media production, virtual assistants, gaming, education, and accessibility technologies. By leveraging deep neural architectures, it offers customizable voice personas with robust emotional nuance and linguistic precision.

Model Input Types

VibeVoice 7B accepts diverse input formats such as plain text, SSML (Speech Synthesis Markup Language) for rich speech control, and prosody parameters to fine-tune intonation, pace, and rhythm. This enables intricate control over voice outputs tailored to different scenarios and user preferences.

Input Length and Context Awareness

The model can process extended conversational inputs while maintaining contextual coherence, making it ideal for dynamic dialogues, narrative storytelling, and multi-turn interactions.

Performance and Output Quality Metrics

Real-Time Speech Generation: Optimized for swift response times, VibeVoice 7B generates high-fidelity speech at near real-time speeds, supporting interactive applications such as live chatbots and virtual personas.
Audio Fidelity: Produces crystal-clear, studio-level voice outputs boasting rich tonal texture, natural prosody, and accurate phonetic detail. The model’s neural vocoder ensures smooth and artifact-free audio synthesis.
Voice Style Variety: Supports multiple voice styles, accents, and emotional tones, from cheerful and energetic to calm and professional, allowing brands to create distinctive auditory identities.

Model Architecture and Innovations

Hybrid Transformer-Based Design: VibeVoice 7B employs a transformer backbone augmented with attention mechanisms specialized for speech features. This hybrid architecture excels in capturing long-range linguistic dependencies and prosodic patterns.
Emotional and Expressive Modulation: Advanced embedding vectors simulate emotional states and speaker intent, enabling expressive speech synthesis that goes beyond monotone robotic voices.
Robust Training Dataset: Trained on a vast multilingual dataset encompassing diverse demographics, accents, and speaking styles, ensuring adaptability across languages and domains.

Core Features and Usage Scenarios

Custom Voice Persona Creation: Users can generate personalized voice variants by adjusting speech style, pitch, and emotional parameters, ideal for interactive voice applications and personalized audio content.
Multi-Domain Applications: Widely used for audiobook narration, voice-overs in videos and commercials, in-game character voices, accessibility tools for the visually impaired, and conversational AI systems.

Use Cases

Interactive Virtual Assistants and Chatbots: Enable AI characters with rich, believable vocal personalities that adjust tone based on conversation flow.
Media and Entertainment Voice Production: Generate diverse character voices and scenarios without costly studio recording sessions.
Accessibility and Assistive Technology: Create natural-sounding screen readers and communication aids supporting emotional expression to enhance user engagement.
Educational Tools: Facilitate language learning and speech therapy applications with clear, expressive pronunciation and customizable pacing.

Use cases

Interactive Virtual Assistants and Chatbots: Enable AI characters with rich, believable vocal personalities that adjust tone based on conversation flow.
Media and Entertainment Voice Production: Generate diverse character voices and scenarios without costly studio recording sessions.
Accessibility and Assistive Technology: Create natural-sounding screen readers and communication aids supporting emotional expression to enhance user engagement.
Educational Tools: Facilitate language learning and speech therapy applications with clear, expressive pronunciation and customizable pacing.

VibeVoice 7B API pricing

VibeVoice 7B pricing in US dollars
Usage	Price	Unit
Speech generation	$0.0007	per second

Prices from the AI/ML API price list, checked on Nov 4, 2026.

VibeVoice 7B vs other speech models: prices

AI/ML API rates for 3 models, per second.

AI/ML API prices in USD per second
Model	Per second
VibeVoice 7B	$0.0007
VibeVoice 1.5B	$0.0007
Seed Audio 1.0	$0.0025

How to use the VibeVoice 7B API

View docs

Build with your agent

Give your agent the model details, code and documentation.

1Get your API key

Create API key

export AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"

2Make your first request

curl 'https://api.aimlapi.com/v1/tts' \
  -H "Authorization: Bearer $AIMLAPI_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
  "model": "vibevoice/7b",
  "script": null
}'

import os
import requests

headers = {
    "Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
    "Content-Type": "application/json",
}
payload = {
    "model": "vibevoice/7b",
    "script": null
}

response = requests.post("https://api.aimlapi.com/v1/tts", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())

const response = await fetch("https://api.aimlapi.com/v1/tts", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "model": "vibevoice/7b",
    "script": null
  }),
});

if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());

All posts

The Most Accurate Speech Models: Pick the Right One
AI News
MiniMax Audio: Voices from China
AI NewsDecember 25, 2024
MAI-Voice-2, MAI-Transcribe 1.5 & MAI-Image 2.5 Guide
AI News

VibeVoice 7B FAQ

What inputs and outputs does VibeVoice 7B support?

VibeVoice 7B takes text as input and returns audio.

How much does VibeVoice 7B cost?

VibeVoice 7B costs $0.0007 per second on AI/ML API, billed as you go with no subscription.

How is VibeVoice 7B billed?

VibeVoice 7B costs $0.0007 per second on AI/ML API, billed as you go with no subscription.

Who created VibeVoice 7B?

VibeVoice 7B was built by Microsoft.

How do I call VibeVoice 7B on AI/ML API?

Send a request to https://api.aimlapi.com/v1/tts with vibevoice/7b as the model id.

Is VibeVoice 7B available through a single API key?

Yes. VibeVoice 7B is served through AI/ML API, so the same key and endpoint format used for other models applies.

Can VibeVoice 7B handle multiple speakers in a single audio generation?

Yes, VibeVoice 7B supports multi-speaker dialogue generation with natural-sounding voices.

Is VibeVoice 7B a good fit for podcast-style or conversational audio content?

It is built for multi-speaker dialogue generation, making it suited to conversational audio such as podcasts.

What kind of use cases is VibeVoice 7B designed for?

It is designed for high-quality text-to-speech with multi-speaker dialogue generation and natural-sounding voices.

What is the VibeVoice 7B model ID?

Use vibevoice/7b as the model parameter. Requests go to https://api.aimlapi.com/v1/tts with your AI/ML API key.

Start building with VibeVoice 7B

Get API key

VibeVoice 7B API

vibevoice/7b

Build with your agent

Give your agent the model details, code and documentation.

1Get your API key

Create API key

export AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"

2Make your first request

curl 'https://api.aimlapi.com/v1/tts' \
  -H "Authorization: Bearer $AIMLAPI_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
  "model": "vibevoice/7b",
  "script": null
}'

import os
import requests

headers = {
    "Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
    "Content-Type": "application/json",
}
payload = {
    "model": "vibevoice/7b",
    "script": null
}

response = requests.post("https://api.aimlapi.com/v1/tts", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())

const response = await fetch("https://api.aimlapi.com/v1/tts", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "model": "vibevoice/7b",
    "script": null
  }),
});

if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());

Docs for AI agents Plain-text index · llms.txt API documentation API reference and examples

Talk to our sales team

Thanks for reaching out

VibeVoice 7B API

VibeVoice 7B overview

VibeVoice 7B by Microsoft

VibeVoice 7B specs

Model Input Types

Input Length and Context Awareness

Performance and Output Quality Metrics

Model Architecture and Innovations

Core Features and Usage Scenarios

Use Cases

Use cases

VibeVoice 7B API pricing

VibeVoice 7B vs other speech models: prices

How to use the VibeVoice 7B API

Build with your agent

1Get your API key

2Make your first request

Related articles

VibeVoice 7B FAQ

Start building with VibeVoice 7B

VibeVoice 7B API

Build with your agent

1Get your API key

2Make your first request

VibeVoice 7B API

VibeVoice 7B overview

VibeVoice 7B by Microsoft

VibeVoice 7B specs

Model Input Types

Input Length and Context Awareness

Performance and Output Quality Metrics

Model Architecture and Innovations

Core Features and Usage Scenarios

Use Cases

Use cases

VibeVoice 7B API pricing

VibeVoice 7B vs other speech models: prices

How to use the VibeVoice 7B API

Build with your agent

1Get your API key

2Make your first request

Models similar to VibeVoice 7B

VibeVoice 1.5B

GPT-4o mini TTS

Seed Audio 1.0

Inworld TTS-1

Inworld TTS-1-Max

Qwen3 TTS Flash

Related articles

The Most Accurate Speech Models: Pick the Right One

MiniMax Audio: Voices from China

MAI-Voice-2, MAI-Transcribe 1.5 & MAI-Image 2.5 Guide

VibeVoice 7B FAQ

Start building with VibeVoice 7B