usamaehsan/whisper-ears
Whisper turbo transcript + audio X-ray in one pass: speaker turns, tone per line, laughs/coughs/sighs with times, 5-clip custom sound detectors, and an LLM-ready report.
Run usamaehsan/whisper-ears 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
|
Speech or any audio (any format ffmpeg reads)
|
|
| language |
string
|
auto
|
Language code like 'en', or 'auto' to detect
|
| task |
None
|
transcribe
|
translate = English words from any language
|
| speakers |
boolean
|
True
|
Label who spoke each line (A, B, ...)
|
| max_speakers |
integer
|
0
Max: 10 |
Most speakers to allow (0 = auto)
|
| sounds |
boolean
|
True
|
Find laughs, sighs, coughs, sneezes, sniffs, throat clearing
|
| custom_examples |
string
|
Optional zip with one folder per sound to find (folder name = label), ~5 short clips each, e.g. doorbell/1.wav. For sounds, not people's voices. Max 10 labels, 20 clips each.
|
|
| initial_prompt |
string
|
Optional spelling hints (names, terms)
|
{
"type": "object",
"title": "Input",
"required": [
"audio"
],
"properties": {
"task": {
"enum": [
"transcribe",
"translate"
],
"type": "string",
"title": "task",
"description": "translate = English words from any language",
"default": "transcribe",
"x-order": 2
},
"audio": {
"type": "string",
"title": "Audio",
"format": "uri",
"x-order": 0,
"description": "Speech or any audio (any format ffmpeg reads)"
},
"sounds": {
"type": "boolean",
"title": "Sounds",
"default": true,
"x-order": 5,
"description": "Find laughs, sighs, coughs, sneezes, sniffs, throat clearing"
},
"language": {
"type": "string",
"title": "Language",
"default": "auto",
"x-order": 1,
"description": "Language code like 'en', or 'auto' to detect"
},
"speakers": {
"type": "boolean",
"title": "Speakers",
"default": true,
"x-order": 3,
"description": "Label who spoke each line (A, B, ...)"
},
"max_speakers": {
"type": "integer",
"title": "Max Speakers",
"default": 0,
"maximum": 10,
"minimum": 0,
"x-order": 4,
"description": "Most speakers to allow (0 = auto)"
},
"initial_prompt": {
"type": "string",
"title": "Initial Prompt",
"x-order": 7,
"nullable": true,
"description": "Optional spelling hints (names, terms)"
},
"custom_examples": {
"type": "string",
"title": "Custom Examples",
"format": "uri",
"x-order": 6,
"nullable": true,
"description": "Optional zip with one folder per sound to find (folder name = label), ~5 short clips each, e.g. doorbell/1.wav. For sounds, not people's voices. Max 10 labels, 20 clips each."
}
}
}
Output schema
The shape of the response you’ll get when you run this model with an API.
{
"title": "Output"
}