sherpa-ncnn is queued for analysis. Crawling candidates and ranking them takes a few minutes. This page updates itself when the list exists — you can also come back later.
What is known so far comes from the other direction:
These projects were analysed and named sherpa-ncnn among their alternatives. The relationship is not symmetric — how sherpa-ncnn rates them is a separate judgement, made when sherpa-ncnn is analysed in its own right.
calls sherpa-ncnn “optimized for ncnn on mobile/embedded, use when low memory and fast inference matter.”
Offline speech recognition API for Android, iOS, Raspberry Pi and servers with Python, Java, C# and Node
calls sherpa-ncnn “Choose for compact real-time ASR on edge devices with Kaldi-trained models and ncnn acceleration, trading Whisper accuracy for lower resources.”
Port of OpenAI's Whisper model in C/C++
calls sherpa-ncnn “Choose if you need real-time on-device ASR with ncnn and Kaldi models, especially for offline mobile/edge use.”
Faster Whisper transcription with CTranslate2
calls sherpa-ncnn “Choose when you need very small-footprint offline ASR/VAD on edge devices using ncnn.”
A PyTorch-based Speech Toolkit
calls sherpa-ncnn “choose for on-device real-time ASR/VAD with ncnn on many platforms.”
Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
calls sherpa-ncnn “Use when you prefer ncnn over ONNX Runtime for some embedded platforms and need only real-time ASR/VAD from the same k2-fsa family.”
Speech-to-text, text-to-speech, speaker diarization, speech enhancement, source separation, and VAD using next-gen Kaldi with onnxruntime without Internet connection. Support embedded systems, Android, iOS, HarmonyOS, Raspberry Pi, RISC-V, RK NPU, Axera NPU, Ascend NPU, x86_64 servers, websocket server/client, support 12 programming languages
calls sherpa-ncnn “choose when you need an ultra-lightweight ncnn-based ASR/VAD for embedded devices and don't need TTS or intent.”
Very low latency speech to text, intent recognition, and text to speech, for building voice agents and interfaces
calls sherpa-ncnn “when targeting embedded CPUs with ncnn and need no diarization”
Real-time, local speech-to-text with streaming ASR, speaker diarization, translation, and OpenAI/Deepgram-compatible APIs.