Geting started¶
Features¶
The Next-gen Kaldi currently supports speech recognition (ASR), speech synthesis (TTS), keyword spotting (KWS), voice activity detection (VAD), speaker identification, spoken language identification, and so on. Some of them provide both training and deployment pipeline, while some are only deployment supports based on excellent third-party open-source libraries. The specific details are as follows:
| Task | Training | Deployment | Docs |
|---|---|---|---|
| Speech recognition(ASR) | Training Deployment | ||
| Speech synthesis(TTS) | Training Deployment | ||
| Keyword spotting(KWS) | Training Deployment | ||
| Voice activity detection(VAD) | Deployment | ||
| Speaker identification | Deployment | ||
| Spoken language identification | Deployment |