Speech and audio
Turn speech into text, turn text into speech, and work with audio.
Start here
The Featured projects with the most downloads.
-
whisper-rs
Rust bindings for whisper.cpp
467k downloads (30 days) Latest release: 0.16.0, 12 Mar 2026
-
sherpa-onnx
Safe Rust wrapper for sherpa-onnx speech recognition toolkit
260k downloads (30 days) Latest release: 2.0.1, 27 Oct 2026
-
nnnoiseless
Audio denoising, derived from Xiph's RNNoise library
173k downloads (30 days) Latest release: 0.5.2, 18 Dec 2025
All projects
Order: strongest tier first, then most downloads, then name, with unmeasured values last. A library counts downloads over 30 days. An app counts binary downloads per release. An app with none counts its crate downloads over 30 days.
What do the tiers mean?
- Featured
- Shown first. It passes every Featured gate.
- Listed
- Shown in the normal listing. No other tier applies.
- New
- A project whose first release is recent.
- Archived
- No longer maintained, or ended by its author.
Tiers come from published rules over raw signals. Stars alone never qualify a project.
Active dependents: crates from other owners that depend on this project in a current release. One owner counts at most once.
Movement needs more history.
| Project | Downloads | Dependents | Stars | Latest release | Runs on (GPU) | Native code | License |
|---|---|---|---|---|---|---|---|
| Featured 4 projects | |||||||
|
whisper-rs Featured
Rust bindings for whisper.cpp |
466,717 downloads over 30 days | 98 | Not measured | 0.16.0 | CUDA, Metal, Vulkan | Binds native code: piper, piper1-rs-sys, whisper, whisper-rs-sys. Build uses a native toolchain: bindgen, cmake | Unlicense |
|
sherpa-onnx Featured
Safe Rust wrapper for sherpa-onnx speech recognition toolkit |
259,599 downloads over 30 days | 28 | 15,834 | 2.0.1 | GPU support unknown | Binds native code: sherpa-onnx, sherpa-onnx-sys | Apache-2.0 |
|
nnnoiseless Featured
Audio denoising, derived from Xiph's RNNoise library |
172,829 downloads over 30 days | 27 | 370 | 0.5.2 | GPU support unknown | Binds native code: hdf5-sys (feature train) | BSD-3-Clause |
|
earshot Featured
Ridiculously fast & accurate voice activity detection in pure Rust |
164,695 downloads over 30 days | 14 | 242 | 1.2.2 | GPU support unknown | No native code found | Apache-2.0 OR MIT |
| Listed 33 projects | |||||||
|
sonora Listed
Automatic Gain Control 2 (AGC2) with RNN VAD for WebRTC Audio Processing |
551,810 downloads over 30 days | 3 | 99 | 0.3.0 | GPU support unknown | No native code found | BSD-3-Clause |
|
transcribe-rs Listed
A simple library to help you transcribe audio |
95,741 downloads over 30 days | 6 | 279 | 0.3.12 | GPU support unknown | Binds native code: ort-sys (feature onnx), whisper-rs-sys (feature whisper-cpp) | MIT |
|
Rust interface for the WebRTC Voice-Activity-Detction Module |
60,022 downloads over 30 days | 7 | 31 | 0.4.0 | GPU support unknown | Build uses a native toolchain: cc | MIT |
|
parakeet-rs Listed
Fast ASR & Speaker Diarization with NVIDIA Parakeet via ONNX |
51,716 downloads over 30 days | 8 | 421 | 0.4.4 | CUDA, WebGPU, Metal, Vulkan | Binds native code: ort-sys (feature api-24) | Apache-2.0 OR MIT |
|
Voice Activity Detector using the Silero VAD model. |
36,149 downloads over 30 days | 11 | 64 | 0.2.1 | GPU support unknown | Binds native code: ort-sys | MIT |
|
mel_spec Listed
Mel spectrograms aligned to the results from the whisper.cpp, pytorch and librosa reference implementations and suited to streaming audio. |
26,437 downloads over 30 days | 6 | 91 | 0.5.0 | CUDA, WebGPU | Binds native code: ort-sys (feature ort-tensor). Build uses a native toolchain: cc | MIT |
|
rustfst Listed
Library for constructing, combining, optimizing, and searching weighted finite-state transducers (FSTs). |
24,531 downloads over 30 days | 10 | 192 | 1.3.1 | GPU support unknown | No native code found | Apache-2.0 OR MIT |
|
tts-rs Listed
High-level Text-To-Speech (TTS) interface |
21,892 downloads over 30 days | 19 | 143 | 0.26.3 | GPU support unknown | Binds native code: speech-dispatcher-sys, tolk-sys (feature tolk) | MIT |
|
jpreprocess Listed
Japanese text preprocessor for Text-to-Speech application (OpenJTalk rewrite in rust language). |
14,491 downloads over 30 days | 7 | 61 | 0.15.0 | GPU support unknown | No native code found | BSD-3-Clause |
|
piper-rs Listed
Use piper TTS models in Rust |
7,898 downloads over 30 days | 9 | 67 | 0.2.0 | GPU support unknown | Binds native code: espeak-rs-sys, sonic-rs-sys, ort-sys. Build uses a native toolchain: bindgen, cc | MIT |
|
Generate music based on natural language prompts using LLMs running locally |
7,531 binary downloads per release | Not applicable | 1,471 | 0.3.27 | CUDA | Binds native code: alsa-sys, openssl-sys, ort-sys | MIT |
|
Safe wrapper around the Vosk API Speech Recognition Toolkit |
7,299 downloads over 30 days | 4 | 182 | 0.3.1 | GPU support unknown | Binds native code: vosk-sys | MIT |
|
msedge-tts Listed
This library is a wrapper of MSEdge Read aloud function API. You can use it to synthesize text to speech with many voices MS provided. |
5,349 downloads over 30 days | 3 | 27 | 0.4.1 | GPU support unknown | No native code found | Apache-2.0 OR MIT |
|
xn-ptts Listed
Phonon text-to-speech inference and serving |
4,761 downloads over 30 days | 1 | 467 | 0.6.0 | CUDA, WebGPU, Metal, Vulkan | No native code found | Apache-2.0 OR MIT |
|
pyannote-rs Listed
Speaker diarization using pyannote in Rust |
4,394 downloads over 30 days | 4 | 131 | 0.3.4 | GPU support unknown | Binds native code: knf-rs-sys, ort-sys | MIT |
|
text-processing-rs Listed
Inverse Text Normalization (ITN) — convert spoken-form ASR output to written form |
3,608 downloads over 30 days | 2 | 48 | 0.2.2 | GPU support unknown | No native code found | Apache-2.0 |
|
any-tts Listed
A Rust TTS library with Candle backends and runtime adapters for modern open TTS models |
3,404 downloads over 30 days | 2 | 23 | 0.2.0 | CUDA, Metal | No native code found | Apache-2.0 OR MIT |
|
Open-source, local-first Granola/Otter alternative that Claude Code, Codex, Cursor, and any MCP client can query. Meetings, calls, and voice memos transcribed on-device into markdown you own. |
3,152 binary downloads per release | Not applicable | 1,595 | 0.32.1 | CUDA, Metal, Vulkan | Binds native code: alsa-sys, libsqlite3-sys, knf-rs-sys (feature diarize), ort-sys (feature diarize), whisper-rs-sys (feature coreml) | MIT |
|
silero Listed
Production-oriented Rust wrapper for the Silero VAD ONNX model. |
1,888 downloads over 30 days | 3 | 6 | 0.7.0 | CUDA, ROCm | Binds native code: ort-sys (feature coreml) | Apache-2.0 OR MIT |
|
moshi Listed
moshi, a real-time voice AI Python bindings |
1,859 downloads over 30 days | 3 | 11,458 | 0.6.4 | CUDA, Metal | Binds native code: alsa-sys, openssl-sys, opusic-sys, sentencepiece-sys | Apache-2.0 OR MIT |
|
Voice-to-text CLI for terminal users |
658 binary downloads per release | Not applicable | 128 | 0.7.2 | GPU support unknown | Binds native code: alsa-sys | MIT |
|
A simple text-to-speech client for Azure TTS API. Python bindings |
488 binary downloads per release | Not applicable | 496 | 6.1.0 | GPU support unknown | No native code found | MIT |
|
scribble Listed
High-level Rust API for audio transcription using Whisper |
457 downloads over 30 days | 0 | 105 | 0.5.4 | CUDA, Metal, Vulkan | Binds native code: whisper-rs-sys | MIT |
|
A Rust implementation of the Qwen3 Text-to-Speech (TTS) model inference. |
204 binary downloads per release | Not applicable | 237 | Not measured | Not measured | None stated | |
|
lele Listed
Bare-metal machine learning inference library in Rust |
132 downloads over 30 days | 1 | 183 | 0.2.0 | GPU support unknown | No native code found | MIT |
|
Voice for the Linux desktop: dictation, read-aloud (TTS), and voice-driven LLM commands, with Groq, OpenAI, Deepgram, and local Whisper backends |
61 binary downloads per release | Not applicable | 149 | 0.2.0 | CUDA, Vulkan | Binds native code: alsa-sys, mac-notification-sys, wayland-sys (feature overlay), whisper-rs-sys (feature local-whisper) | MIT |
|
gradbot Listed
Open source framework to vibecode and prototype voice agents with Gradium APIs |
31 downloads over 30 days | 3 | 152 | 0.6.4 | GPU support unknown | No native code found | Apache-2.0 OR MIT |
|
qwen3-asr-rs Listed
Pure Rust implementation of Qwen3 ASR (Automatic Speech Recognition) with libtorch and MLX backends |
21 downloads over 30 days | 0 | 254 | 0.2.0 | CUDA | Binds native code: torch-sys (feature tch-backend). Build uses a native toolchain: cmake | Apache-2.0 |
|
Inference server for LLMs, speech and embeddings: OpenAI-compatible routes with streaming, plus speech synthesis (TTS) and transcription (ASR). Serves Llama, Mistral, Qwen, Mamba2, MLA+MoE and hybrid models from SafeTensors, AWQ, GPTQ and GGUF. |
8 binary downloads per release | Not applicable | 7 | 0.7.1 | CUDA, WebGPU | Binds native code: ring (feature distributed) | Apache-2.0 |
|
kitten-tts-rs Listed
Rust port of KittenTTS — ultra-lightweight ONNX-based text-to-speech |
3 downloads over 30 days | 0 | 314 | 0.2.2 | CUDA | Binds native code: audiopus_sys, mp3lame-sys, ort-sys | Apache-2.0 |
|
cohere-transcribe-rs Listed
Rust implementation for the CohereLabs/cohere-transcribe-03-2026 speech-to-text model. Includes a CLI and an OpenAI-compatible API server. |
0 downloads over 30 days | 0 | 101 | 0.1.1 | CUDA | Binds native code: torch-sys (feature tch-backend). Build uses a native toolchain: cmake | Apache-2.0 |
|
Hibiki is a model for streaming speech translation (also known as simultaneous translation). |
0 downloads over 30 days | Not applicable | 1,537 | 0.1.2 | CUDA, Metal | Binds native code: sentencepiece-sys | Apache-2.0 OR MIT |
|
YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context |
0 binary downloads per release | Not applicable | 22,279 | Not measured | Not measured | Not measured | |
| New 3 projects | |||||||
|
transcribe-cpp New
Rust bindings for transcribe.cpp, a C/C++ speech-to-text library built on ggml |
50,251 downloads over 30 days | 3 | 2,149 | 0.4.1 | CUDA, Metal, ROCm, Vulkan | Binds native code: transcribe, transcribe-cpp-sys, transcribe_cpp. Build uses a native toolchain: cmake | MIT |
|
voxr New
Audio and speech on boostr: compressed-audio decode, resampling, DSP and grapheme to phoneme, with speech synthesis (Kokoro, VITS), voice cloning (VoxCPM2), transcription with word timings (Whisper), voice activity (Silero) and audio-text scoring (CLAP), on CPU or CUDA. |
642 downloads over 30 days | 2 | 2 | 0.9.1 | CUDA | No native code found | Apache-2.0 |
|
flowcat New
Self-hosted native-Rust real-time voice-agent runtime — one binary, no Python or softswitch: transports, serializers, codec, and a speech-to-speech (Gemini Live) pipeline. |
46 downloads over 30 days | 0 | 127 | 0.1.0 | GPU support unknown | Binds native code: audiopus_sys (feature local), ort-sys (feature vad-ort), whisper-rs-sys (feature stt-whisper-local) | Apache-2.0 |
| Archived 1 project | |||||||
|
sherpa-rs Archived
Rust bindings to https://github.com/k2-fsa/sherpa-onnx |
9,888 downloads over 30 days | 4 | 313 | 0.6.8 | CUDA | Binds native code: sherpa-rs-sys. Build uses a native toolchain: bindgen, cmake | MIT |