zsxkib / voxtral
Voxtral Mini (3B) + Small (24B)🎙️ Speech transcription and audio understanding in 8 languages🧠
Run zsxkib/voxtral with an API
Use one of our client libraries to get started quickly. Clicking on a library will take you to the Playground tab where you can tweak different inputs, see the results, and copy the corresponding code to use in your own project.
Input schema
The fields you can use to run this model with an API. If you don't give a value for a field its default value will be used.
| Field | Type | Default value | Description |
|---|---|---|---|
| audio |
string
|
Audio file to process.
|
|
| mode |
string
(enum)
|
transcription
Options: transcription, understanding |
Choose processing mode: 'transcription' converts speech to text, 'understanding' analyzes audio content using prompts.
|
| prompt |
string
|
What can you tell me about this audio?
|
Question or instruction for understanding mode (e.g., 'What is the speaker discussing?', 'Summarize this audio'). Ignored in transcription mode.
|
| language |
string
(enum)
|
Auto-detect
Options: Auto-detect, English, French, German, Spanish, Italian, Portuguese, Dutch, Russian, Chinese, Japanese, Arabic |
Audio language. 'Auto-detect' works for most content, or choose a specific language for better accuracy.
|
| model_size |
string
(enum)
|
mini
Options: mini, small |
Model selection: 'mini' (3B) is faster and uses less GPU memory, 'small' (24B) provides higher accuracy for complex audio.
|
| max_tokens |
integer
|
500
Min: 50 Max: 1000 |
Maximum response length. Higher values allow longer outputs but increase processing time.
|
{
"type": "object",
"title": "Input",
"required": [
"audio"
],
"properties": {
"mode": {
"enum": [
"transcription",
"understanding"
],
"type": "string",
"title": "mode",
"description": "Choose processing mode: 'transcription' converts speech to text, 'understanding' analyzes audio content using prompts.",
"default": "transcription",
"x-order": 1
},
"audio": {
"type": "string",
"title": "Audio",
"format": "uri",
"x-order": 0,
"description": "Audio file to process."
},
"prompt": {
"type": "string",
"title": "Prompt",
"default": "What can you tell me about this audio?",
"x-order": 2,
"description": "Question or instruction for understanding mode (e.g., 'What is the speaker discussing?', 'Summarize this audio'). Ignored in transcription mode."
},
"language": {
"enum": [
"Auto-detect",
"English",
"French",
"German",
"Spanish",
"Italian",
"Portuguese",
"Dutch",
"Russian",
"Chinese",
"Japanese",
"Arabic"
],
"type": "string",
"title": "language",
"description": "Audio language. 'Auto-detect' works for most content, or choose a specific language for better accuracy.",
"default": "Auto-detect",
"x-order": 3
},
"max_tokens": {
"type": "integer",
"title": "Max Tokens",
"default": 500,
"maximum": 1000,
"minimum": 50,
"x-order": 5,
"description": "Maximum response length. Higher values allow longer outputs but increase processing time."
},
"model_size": {
"enum": [
"mini",
"small"
],
"type": "string",
"title": "model_size",
"description": "Model selection: 'mini' (3B) is faster and uses less GPU memory, 'small' (24B) provides higher accuracy for complex audio.",
"default": "mini",
"x-order": 4
}
}
}
Output schema
The shape of the response you’ll get when you run this model with an API.
{
"type": "string",
"title": "Output"
}