NVIDIA NeMo
NVIDIA framework for speech and language models
NVIDIA NeMo is an open-source framework for building speech and language models. Its speech toolkit includes recognition models such as Parakeet and Canary, along with training recipes.
More in Speech-to-text
Featured
Deepgram
Speech-to-text
Speech-to-text and text-to-speech APIs
FeaturedOpenAI Whisper
Speech-to-text
Open-source multilingual speech recognition
Amazon Transcribe
Speech-to-text
AWS speech-to-text service
AssemblyAI
Speech-to-text
Speech-to-text API with audio intelligence
faster-whisper
Speech-to-text
Whisper reimplemented with CTranslate2
Gladia
Speech-to-text
Speech-to-text API for files and live audio