Unsigned
7
1
voice-stt
:Use the Pull Tag button to download this ModelKit.
Or, read our KitOps documentation to learn how to use kit unpack --filter to download only the components you need.
ModelKit Tag Metadata
- Author
- Jozu AI
- Date added
- Size
- 486.2MB
- Digest
- Total pulls
- 7
Package
- Name
- voice-stt-faster-whisper
- Version
- 1.0.0
- Authors
- Jozu AI
- Description
- Faster-Whisper Small model for real-time speech-to-text in telephony pipelines. Optimized for 8kHz/16kHz audio with VAD filtering for natural conversation flow.
Model
- Name
- faster-whisper-small
- Path
- ./weights/
- License
- MIT
- Parts
- N/A
- Parameters
{
"language": "en",
"beam_size": 3,
"vad_filter": true,
"compute_type": "int8",
"vad_parameters": {
"min_silence_duration_ms": 500
}
}