Audio Hub › Speech-to-text

CrispASR

C++ ggml runtime hub for multilingual ASR and TTS models: Cohere Transcribe, Parakeet TDT, Voxtral, Canary 1B v2, etc, plus universal forced alignment, and more

Category
Speech-to-text
Type
Open source
Platform
C++ project
Pricing
Free / open source
License
MIT
GitHub stars
540
Last updated
2026-08-13

GitHub repo →

Need the recording as text?

Whatever you record or edit with CrispASR, Whipscribe turns it into an accurate, speaker-labeled transcript — 100+ languages, SRT/VTT/DOCX export, private self-hosted Whisper. About $2 per audio hour, 30 minutes free daily.

Transcribe a file →

Top CrispASR alternatives

transformers🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, fwhisper.cppPort of OpenAI's Whisper model in C/C++voiceboxThe open-source AI voice studio. Clone, dictate, create.HandyA free, open source, and extensible speech-to-text application that works completely offline.meetilyPrivacy first, AI meeting assistant with 4x faster Parakeet/Whisper live transcription, speaker diarization, and Ollama summarization built llamafileDistribute and run LLMs with a single file.

See the full list of CrispASR alternatives →