# espressif/esp-sr

Speech recognition

Repository: https://github.com/espressif/esp-sr
Canonical: https://ross.abutalabs.com/products/esp-sr
Language: C
License: NOASSERTION
License Family: other
Last push: 2026-08-19T12:02:41+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 98, release rhythm 40, longevity 100
- inputs: {"age_days": 2592, "days_push": 14, "days_rel": 559, "gap_med": 14, "n_releases_24m": 2}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1492, forks 224 (observed 2026-08-28T04:04:52.685277+00:00)

## What it is
ESP-SR is Espressif's speech recognition framework for ESP32-series chips, providing wake word detection (WakeNet), voice activity detection (VADNet), speech command recognition (MultiNet), audio front-end processing, and speech synthesis. It ships as an installable component so embedded developers can add offline voice AI to their firmware with minimal effort.

## Use cases
- add a custom wake word to an ESP32-S3 device
- recognize offline voice commands on a microcontroller
- build a voice-controlled smart home device with ESP32
- run voice activity detection on embedded hardware
- add text-to-speech output to an ESP32 project
- do acoustic echo cancellation for full-duplex voice on ESP32
- detect sound direction of arrival with a microphone array

## When to choose
- you are building voice interfaces on ESP32/ESP32-S3/C-series chips
- you need fully offline wake word and command recognition on-device
- you want an integrated pipeline (AFE, VAD, wake word, commands, TTS) as a single component

## When to avoid
- you need speech recognition on servers, desktops, or non-Espressif hardware
- you need large-vocabulary or free-form dictation rather than wake words and fixed commands
- you require an OSI-approved open-source license, since the licensing is custom

## Facets
- artifact type: library
- maturity: active
- function: speech-recognition, tts, audio-processing, machine-learning, sdk
- domain: speech-processing, embedded-systems, artificial-intelligence, iot
- platform: embedded, iot, c, cpp
- tags: wake-word, voice-activity-detection, esp32, espressif, offline-speech, embedded-ai, audio-front-end

## Member repositories
- espressif/esp-sr (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:52.685277+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:33:29.547242+00:00, confidence not recorded.
  - readme: https://github.com/espressif/esp-sr (fetched 2026-08-28T04:04:52.685277+00:00, sha c25011e855a1)
- Data as of 2026-08-30T08:39:29.467469+00:00.