​
What can you do with it?

The /audio-generate command enables you to convert text to speech using the advanced Minimax Speech-02-Turbo model. You can create professional voiceovers, generate multilingual audio content, produce podcasts, create audiobooks, develop voice assistants, and generate high-quality speech with extensive control over voice characteristics, emotions, and audio quality.

​
How to use it?

​
Basic Command Structure

/audio-generate [text]

​
Parameters

Required:
  • text - Text to convert to speech (max 5000 characters). Use <#x#> for pause control (0.01-99.99s)
Optional:
  • pitch - Speech pitch: -12 to 12 (defaults to 0)
  • speed - Speech speed: 0.5 to 2 (defaults to 1)
  • volume - Speech volume: 0 to 10 (defaults to 1)
  • bitrate - Audio bitrate: 32000, 64000, 128000, 256000 (defaults to 128000)
  • channel - Audio channels: “mono”, “stereo” (defaults to “mono”)
  • emotion - Speech emotion: “auto”, “neutral”, “happy”, “sad”, “angry”, “fearful”, “disgusted”, “surprised” (defaults to “auto”)
  • voice_id - Voice selection (defaults to “Wise_Woman”). See available voices below
  • sample_rate - Sample rate: 8000, 16000, 22050, 24000, 32000, 44100 (defaults to 32000)
  • language_boost - Language enhancement (defaults to “None”). See language options below
  • english_normalization - Enable English text normalization for better number reading (boolean, defaults to false)
  • fileLinksExpireInDays - How long generated files remain accessible: 1-7 days (defaults to 7)
  • fileLinksExpireInMinutes - How long generated files remain accessible in minutes (takes precedence over days)

​
Response Format

The command returns: 
{
  "output": [
    {
      "url": "https://generated-file-url",
      "mimeType": "audio/wav"
    }
  ]
}
Note: All generated audio files are automatically saved to your “Multimedia Artifact” file store and will be accessible for the duration specified by the fileLinksExpireInDays parameter.

​
Examples

​
Basic Text-to-Speech

/audio-generate
text: Welcome to our customer service. How can I help you today?
Generates basic speech with default voice and settings.

​
Professional Voiceover

/audio-generate
text: Introducing our revolutionary new product that will change the way you work forever.
voice_id: Deep_Voice_Man
emotion: neutral
pitch: -2
speed: 0.9
volume: 8
bitrate: 256000
sample_rate: 44100
Creates professional male voiceover with deep voice and high-quality audio.

​
Interactive Voice Response (IVR)

/audio-generate
text: Press 1 for sales, <#2#> press 2 for support, <#2#> or stay on the line for an operator.
voice_id: Friendly_Person
emotion: happy
speed: 1.0
channel: mono
english_normalization: true
Generates IVR prompts with pauses and number normalization for better clarity.

​
Multilingual Content

/audio-generate
text: Bonjour et bienvenue dans notre magasin. Comment puis-je vous aider aujourd'hui?
voice_id: Elegant_Man
language_boost: French
emotion: neutral
pitch: 1
speed: 1.1
Creates French speech with language-specific enhancements.

​
Podcast Introduction

/audio-generate
text: Welcome to Tech Talk Weekly, <#1#> the podcast where we dive deep into the latest technology trends and innovations.
voice_id: Casual_Guy
emotion: happy
pitch: 0
speed: 1.0
volume: 7
bitrate: 128000
channel: stereo
sample_rate: 44100
Generates engaging podcast intro with stereo audio and natural pacing.

​
Audiobook Narration

/audio-generate
text: Chapter One: The Journey Begins. <#3#> It was a dark and stormy night when Sarah first discovered the mysterious letter hidden in her grandmother's attic.
voice_id: Wise_Woman
emotion: neutral
pitch: 0
speed: 0.8
volume: 6
bitrate: 256000
fileLinksExpireInDays: 7
Creates audiobook-style narration with slower speed and high bitrate for quality.

​
Children’s Content

/audio-generate
text: Once upon a time, in a magical forest, there lived a friendly dragon named Sparkles who loved to help everyone!
voice_id: Lively_Girl
emotion: happy
pitch: 3
speed: 1.2
volume: 8
channel: mono
Generates cheerful children’s content with animated voice characteristics.

​
Corporate Training

/audio-generate
text: In this module, you will learn about workplace safety procedures. <#2#> Please pay careful attention to the following guidelines.
voice_id: Patient_Man
emotion: neutral
pitch: -1
speed: 0.9
volume: 7
english_normalization: true
bitrate: 128000
Creates professional training audio with clear, measured delivery.

​
Emergency Announcement

/audio-generate
text: Attention all employees. <#1#> This is an important safety announcement. Please proceed to the nearest exit in an orderly fashion.
voice_id: Imposing_Manner
emotion: neutral
pitch: -3
speed: 0.8
volume: 10
channel: mono
sample_rate: 32000
Generates clear, authoritative emergency announcement with maximum volume.

​
Marketing Advertisement

/audio-generate
text: Don't miss our incredible summer sale! <#1#> Save up to 50% on all items this weekend only!
voice_id: Exuberant_Girl
emotion: happy
pitch: 2
speed: 1.3
volume: 9
bitrate: 256000
channel: stereo
Creates energetic marketing audio with enthusiastic delivery.

​
Meditation and Wellness

/audio-generate
text: Take a deep breath <#3#> and slowly exhale. <#3#> Feel your body relax as you release all tension.
voice_id: Calm_Woman
emotion: neutral
pitch: -1
speed: 0.6
volume: 5
bitrate: 128000
sample_rate: 44100
Generates soothing meditation audio with slow, calming delivery.

​
Notes

Model Capabilities:
  • High-quality neural text-to-speech synthesis
  • 17 different voice personalities
  • Multilingual support with language-specific enhancements
  • Precise emotion control for natural-sounding speech
  • Advanced pause control with <#x#> notation
  • Professional audio quality up to 44.1kHz/256kbps
Available Voice IDs:
  • Wise_Woman - Mature, knowledgeable female voice
  • Friendly_Person - Warm, approachable neutral voice
  • Inspirational_girl - Uplifting, motivational young female
  • Deep_Voice_Man - Rich, authoritative male voice
  • Calm_Woman - Soothing, peaceful female voice
  • Casual_Guy - Relaxed, conversational male voice
  • Lively_Girl - Energetic, animated young female
  • Patient_Man - Steady, educational male voice
  • Young_Knight - Noble, heroic male voice
  • Determined_Man - Confident, resolute male voice
  • Lovely_Girl - Sweet, gentle female voice
  • Decent_Boy - Polite, well-mannered male voice
  • Imposing_Manner - Authoritative, commanding voice
  • Elegant_Man - Refined, sophisticated male voice
  • Abbess - Dignified, spiritual female voice
  • Sweet_Girl_2 - Charming, endearing female voice
  • Exuberant_Girl - Enthusiastic, spirited female voice
Language Enhancement Options:
  • None - No language-specific processing
  • Automatic - Auto-detect and enhance
  • Chinese - Mandarin Chinese enhancement
  • Chinese,Yue - Cantonese Chinese enhancement
  • English - English language enhancement
  • Arabic, Russian, Spanish, French, Portuguese, German, Turkish, Dutch, Ukrainian, Vietnamese, Indonesian, Japanese, Italian, Korean, Thai, Polish, Romanian, Greek, Czech, Finnish, Hindi
Best Practices:
  • Use <#x#> for precise pause control (e.g., <#1.5#> for 1.5 second pause)
  • Enable english_normalization for better number and abbreviation reading
  • Use higher bitrates (256000) for professional applications
  • Choose appropriate voice_id based on content type and target audience
  • Adjust speed based on content complexity (slower for educational, faster for energetic content)
Limitations:
  • Maximum 5000 characters per request
  • Processing time increases with text length and quality settings
  • Some voices may be more suitable for specific languages
  • Pause control syntax must be exact: <#number#>

​
Model Parameters (minimax/speech-02-turbo)

​
Text-to-Speech Parameters

  • text (required): Text to convert to speech (max 5000 chars). Use <#x#> for pause control (0.01-99.99s)
  • pitch: Speech pitch (-12 to 12, default: 0)
  • speed: Speech speed (0.5 to 2, default: 1)
  • volume: Speech volume (0 to 10, default: 1)
  • bitrate: Bitrate (32000, 64000, 128000, 256000, default: 128000)
  • channel: Audio channels (“mono”, “stereo”, default: “mono”)
  • emotion: Speech emotion (“auto”, “neutral”, “happy”, “sad”, “angry”, “fearful”, “disgusted”, “surprised”, default: “auto”)
  • voice_id: Voice ID (default: “Wise_Woman”). See available voices above
  • sample_rate: Sample rate (8000, 16000, 22050, 24000, 32000, 44100, default: 32000)
  • language_boost: Language enhancement (default: “None”). See language options above
  • english_normalization: Enable English text normalization for better number reading (boolean, default: false, slightly increases latency)