vsref
sherpa-onnx logo

sherpa-onnx Review

OSS

On-device ASR/TTS runtime for edge and embedded

See pricing

Facts verified Jul 20, 2026Website →

What we know about sherpa-onnx

This is our verified profile of sherpa-onnx, a speech-to-text apis platform - on-device ASR/TTS runtime for edge and embedded. Every fact about sherpa-onnx below carries the source it came from and the day we checked it.

sherpa-onnx does not publish a public per-minute rate we have been able to verify, so the price figure here stays blank until we can confirm one. We would rather show nothing than a guessed number; if you have a current sherpa-onnx quote, it is the fastest way for us to close that gap.

On capabilities, sherpa-onnx covers word-level timestamps, custom vocabulary / keyterm boosting, self-host / on-prem option, and websocket streaming api, and does not offer entity detection, sentiment analysis, and summarization endpoint. Each of those is verified against sherpa-onnx's own docs or dashboard, not marketing copy.

For compliance, with sherpa-onnx: SOC 2 Type II is not reported. If you are in a regulated space, confirm the current posture with sherpa-onnx before you commit, since these change plan by plan.

In total we track 19 verified facts for sherpa-onnx today, each linking the primary source it came from so you can check our work - and vendor claims we have not measured ourselves are labeled as such on the sherpa-onnx fact sheet below.

Reviewed by vsref Editorialfacts verified Jul 20, 2026Methodology →

Fact sheet

Capabilities
Capabilities facts
Speaker diarization✓ IncludedJul 20
PII redaction✗ Not availableJul 20
Word-level timestamps✓ YesJul 20
Custom vocabulary / keyterm boosting✓ YesJul 20
Entity detection✗ NoJul 20
Sentiment analysis✗ NoJul 20
Summarization endpoint✗ NoJul 20
Speech translation✗ NoJul 20
Priced audio-intelligence add-onsVAD, punctuation, audio tagging, keyword spotting, lang IDJul 20
Compliance & trust
Compliance & trust facts
HIPAA BAA available✗ Not availableJul 20
SOC 2 Type II✗ NoJul 20
Self-host / on-prem option✓ YesJul 20
Model weights licenseApache-2.0Jul 20
Build experience
Build experience facts
Official SDKsPython, C++, C, C#, Go, Java, Kotlin, Swift, Rust, Dart, JSJul 20
Websocket streaming API✓ YesJul 20
Commercial
Commercial facts
Hardware to self-hostCPU; also GPU/NPU (Jetson, Qualcomm, RKNN, Ascend)Jul 20
Hosted API availableNo - self-hosted runtime library onlyJul 20
Project maintenance statusActive - v1.13.4 released 2026-07Jul 20
GitHub stars13,670 starsJul 20

Considering a switch? Best sherpa-onnx alternatives →

Voice agents in this stack

The engine is one layer: a voice agent hears through its transcription engine, but orchestration, telephony, and turn-taking come from the agent platform. Compare the platforms builders pair sherpa-onnx with: Pipecat, OpenAI Realtime API, Twilio ConversationRelay, or the full voice-agent comparison →

Want it done for you?

sherpa-onnx gets you a transcript; the bot that joins the call, labels speakers, and writes the summary is still your build. If that is more pipeline than you want to own, AI meeting notetakers do the whole job end to end: Otter.ai, Fireflies.ai, Fathom, or the full notetaker comparison.