Temi
Affordable automated transcription service with quick turnaround and easy exports.
Sorting
NVIDIA NeMo ASR
Open toolkit with SOTA ASR models (Conformer, CTC/Transducer) for training and deployment.
Vosk
Lightweight open-source offline ASR toolkit supporting many languages and on-device use.
Deepgram Speech-to-Text
Real-time and batch API with low latency, enterprise scaling, and model choices for accuracy or speed.
Picovoice Leopard
Private, on-device speech-to-text SDK delivering cloud-level accuracy without sending data out.