Docs
Roadmap
Roadmap
We disclose part of our roadmap in this section. We have a good-faith intent to deliver on this roadmap but this intent does not imply a contractual commitment.
⏳ indicates active development with a less than 3 month expected release date.
| Open-source SDK | Pro SDK | Pro is planned to be | |
|---|---|---|---|
| WhisperKit Features | |||
| File Transcription | ⏳ | ~2x faster | |
| Language Detection | |||
| Word Timestamps | |||
| Custom Keywords | |||
| SRT & VTT Output Formats | |||
| Real-time Transcription | ⏳ | ~2x faster | |
| Fast Model Load | |||
| SpeakerKit Features | |||
| Voice Activity Detection | |||
| Speaker Diarization | ⏳ | higher accuracy | |
| RTTM Output Format | |||
| Diarized Transcription | ⏳ | higher accuracy | |
| Speaker identification | ⏳ |
We are also working on on-device inference kits for language models and text-to-speech-models.