Incredibly Fast Whisper
Transcribe audio files with high speed using an optimized Whisper model. Supports multiple languages, speaker diarization, and word-level timestamps.
Quick Start
Requires an each::labs API key. Get one at eachlabs.ai.
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "incredibly-fast-whisper",
"version": "0.0.1",
"input": {
"audio": "https://example.com/podcast-episode.wav",
"language": "english",
"task": "transcribe",
"timestamp": "chunk"
}
}'
Parameters
| Parameter |
Type |
Default |
Description |
| audio |
string |
- |
Audio file to transcribe |
| batch_size |
integer |
24 |
Batch size for processing |
| diarise_audio |
boolean |
false |
Split conversation into segments by speaker |
| hf_token |
string |
- |
HuggingFace API token for accessing private resources |
| language |
string |
None |
Input language. enum: None (auto-detect), afrikaans, amharic, arabic, azerbaijani, belarusian, bosnian, breton, bulgarian, cantonese, catalan, chinese, croatian, czech, danish, dutch, english, estonian, finnish, french, german, hebrew, hindi, hungarian, italian, japanese, korean, lithuanian, macedonian, mongolian, myanmar, nepali, norwegian, polish, portuguese, romanian, russian, serbian, slovak, slovenian, spanish, swahili, swedish, tatar, telugu, turkish, ukrainian, welsh |
| task |
string |
transcribe |
Task to perform. enum: transcribe |
| timestamp |
string |
chunk |
Timestamp granularity. enum: chunk, word |
Examples
Transcribe with speaker diarization:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "incredibly-fast-whisper",
"version": "0.0.1",
"input": {
"audio": "https://example.com/interview.wav",
"language": "english",
"diarise_audio": true,
"timestamp": "word",
"batch_size": 24
}
}'
Auto-detect language transcription:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "incredibly-fast-whisper",
"version": "0.0.1",
"input": {
"audio": "https://example.com/foreign-audio.mp3",
"language": "None",
"task": "transcribe",
"timestamp": "chunk"
}
}'
Related Models
Documentation
1---2name: incredibly-fast-whisper3description: Incredibly Fast Whisper | Audio Transcription. Transcribe audio files quickly with speaker diarization and multi-language support. Triggers: whisper, transcription, audio to text, speech to text, transcribe audio4---56# Incredibly Fast Whisper78Transcribe audio files with high speed using an optimized Whisper model. Supports multiple languages, speaker diarization, and word-level timestamps.910## Quick Start1112> Requires an each::labs API key. Get one at [eachlabs.ai](https://eachlabs.ai).1314```bash15curl -X POST https://api.eachlabs.ai/v1/prediction \16 -H "Content-Type: application/json" \17 -H "X-API-Key: $EACHLABS_API_KEY" \18 -d '{19 "model": "incredibly-fast-whisper",20 "version": "0.0.1",21 "input": {22 "audio": "https://example.com/podcast-episode.wav",23 "language": "english",24 "task": "transcribe",25 "timestamp": "chunk"26 }27 }'28```2930## Parameters3132| Parameter | Type | Default | Description |33|-----------|------|---------|-------------|34| audio | string | - | Audio file to transcribe |35| batch_size | integer | 24 | Batch size for processing |36| diarise_audio | boolean | false | Split conversation into segments by speaker |37| hf_token | string | - | HuggingFace API token for accessing private resources |38| language | string | None | Input language. enum: None (auto-detect), afrikaans, amharic, arabic, azerbaijani, belarusian, bosnian, breton, bulgarian, cantonese, catalan, chinese, croatian, czech, danish, dutch, english, estonian, finnish, french, german, hebrew, hindi, hungarian, italian, japanese, korean, lithuanian, macedonian, mongolian, myanmar, nepali, norwegian, polish, portuguese, romanian, russian, serbian, slovak, slovenian, spanish, swahili, swedish, tatar, telugu, turkish, ukrainian, welsh |39| task | string | transcribe | Task to perform. enum: transcribe |40| timestamp | string | chunk | Timestamp granularity. enum: chunk, word |4142## Examples4344**Transcribe with speaker diarization:**45```bash46curl -X POST https://api.eachlabs.ai/v1/prediction \47 -H "Content-Type: application/json" \48 -H "X-API-Key: $EACHLABS_API_KEY" \49 -d '{50 "model": "incredibly-fast-whisper",51 "version": "0.0.1",52 "input": {53 "audio": "https://example.com/interview.wav",54 "language": "english",55 "diarise_audio": true,56 "timestamp": "word",57 "batch_size": 2458 }59 }'60```6162**Auto-detect language transcription:**63```bash64curl -X POST https://api.eachlabs.ai/v1/prediction \65 -H "Content-Type: application/json" \66 -H "X-API-Key: $EACHLABS_API_KEY" \67 -d '{68 "model": "incredibly-fast-whisper",69 "version": "0.0.1",70 "input": {71 "audio": "https://example.com/foreign-audio.mp3",72 "language": "None",73 "task": "transcribe",74 "timestamp": "chunk"75 }76 }'77```7879## Related Models8081- [youtube-transcriptor](../youtube-transcriptor/) - YouTube video transcription82- [xtts-v2](../xtts-v2/) - Text-to-speech synthesis83- [openvoice](../openvoice/) - Voice cloning and speech8485## Documentation8687- [each::labs Docs](https://docs.eachlabs.ai)88- [API Reference](https://docs.eachlabs.ai/api/overview)
awesome-genmedia