Add ten-vad Rust API example to remove silences from a file by sportiz91 · Pull Request #3778 · k2-fsa/sherpa-onnx

sportiz91 · 2026-07-21T16:39:47Z

Part of #3210 ("Add ten-vad example to remove silences from a file").

This ports rust-api-examples/examples/silero_vad_remove_silence.rs to ten-vad, following the existing Java (TenVadRemoveSilence.java) and Pascal (remove_silence_ten_vad.pas) examples:

examples/ten_vad_remove_silence.rs — uses TenVadModelConfig with an explicit window_size = 256 (matching the other language examples) and feeds the VAD in 256-sample chunks.
run-ten-vad-remove-silence.sh — downloads ten-vad.onnx and the test wave file if needed, then runs the example.
README.md — added example 49 to the table and the "Run it" section.
.github/scripts/test-rust.sh — runs the new script right after the silero VAD one.

Tested on Linux x86_64 (Rust 1.94, default static linking):

Input WAV: sample rate: 16000, num samples: 4358370, duration: 272.40s
Saved speech-only audio to ./no-silence-ten-vad.wav

=== Summary ===
Input:  sample rate = 16000, samples = 4358370, duration = 272.40s
Output: sample rate = 16000, samples = 2513376, duration = 157.09s
Removed non-speech: 42.33% of input removed

Summary by CodeRabbit

New Features
- Added a Rust CLI example to remove non-speech segments from WAV files using ten-vad, writing a speech-only output WAV and printing a processing summary.
- Added a convenience script to fetch the required model and input audio (when missing) and run the example.
Documentation
- Updated the Rust API examples README with “Example 49” and the command to execute the new silence-removal workflow.
Tests
- Included the new silence-removal run script in the Rust test/run sequence.

coderabbitai · 2026-07-21T16:40:11Z

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 67d74fdd-b23c-45ec-8af4-f9b6e34f2ea1

📥 Commits

Reviewing files that changed from the base of the PR and between 4a28f0c and beffc47.

📒 Files selected for processing (2)

rust-api-examples/examples/ten_vad_remove_silence.rs
rust-api-examples/run-ten-vad-remove-silence.sh

🚧 Files skipped from review as they are similar to previous changes (2)

rust-api-examples/run-ten-vad-remove-silence.sh
rust-api-examples/examples/ten_vad_remove_silence.rs

📝 Walkthrough

Walkthrough

Adds a Rust ten-vad example that removes non-speech audio from WAV files, a script that downloads required assets and runs it, README documentation, and integration into the Rust test sequence.

Changes

Ten-VAD silence removal

Layer / File(s)	Summary
Rust VAD processing pipeline `rust-api-examples/examples/ten_vad_remove_silence.rs`	Adds CLI arguments, ten-vad configuration, chunked waveform processing, speech-segment collection, filtered WAV output, and duration reporting.
Example runner and asset setup `rust-api-examples/run-ten-vad-remove-silence.sh`	Downloads `ten-vad.onnx` and `lei-jun-test.wav` when missing, then runs the Rust example with the corresponding paths.
Documentation and test registration `rust-api-examples/README.md`, `.github/scripts/test-rust.sh`	Documents Example 49 and adds the runner to the Rust test sequence.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Sequence Diagram(s)

sequenceDiagram
  participant Runner
  participant RustExample
  participant VoiceActivityDetector
  participant OutputWAV
  Runner->>RustExample: provide input, output, and model paths
  RustExample->>VoiceActivityDetector: accept_waveform audio chunks
  VoiceActivityDetector-->>RustExample: return detected speech segments
  RustExample->>VoiceActivityDetector: flush remaining audio
  VoiceActivityDetector-->>RustExample: return remaining speech segments
  RustExample->>OutputWAV: write speech-only samples

Possibly related PRs

k2-fsa/sherpa-onnx#3209: Extends .github/scripts/test-rust.sh with additional Rust example runner invocations.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly matches the main change: a new ten-vad Rust API example for removing silence from audio files.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

gemini-code-assist

Code Review

This pull request introduces a new Rust API example and script for removing non-speech segments from audio files using ten-vad. Feedback on the changes includes validating that the input WAV file has a sample rate of 16000 Hz to prevent abrupt process termination, pre-allocating capacity for the speech samples vector to optimize performance, and adding the -f flag to curl commands in the shell script to ensure failures are handled gracefully.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

gemini-code-assist · 2026-07-21T16:41:39Z

+    println!(
+        "Input WAV: sample rate: {}, num samples: {}, duration: {:.2}s",
+        sample_rate, input_num_samples, input_duration
+    );


Since ten-vad strictly requires a sample rate of 16000 Hz (and the underlying C++ implementation will abruptly terminate the process with exit(-1) if it is not), it is highly recommended to validate the sample rate in Rust and return a graceful error.

Suggested change

println!(

"Input WAV: sample rate: {}, num samples: {}, duration: {:.2}s",

sample_rate, input_num_samples, input_duration

);

println!(

"Input WAV: sample rate: {}, num samples: {}, duration: {:.2}s",

sample_rate, input_num_samples, input_duration

);

if sample_rate != 16000 {

anyhow::bail!(

"ten-vad expects a sample rate of 16000 Hz, but the input file has {} Hz",

sample_rate

);

}

gemini-code-assist · 2026-07-21T16:41:39Z

+    let vad = VoiceActivityDetector::create(&vad_config, 30.0)
+        .ok_or_else(|| anyhow::anyhow!("Failed to create VoiceActivityDetector"))?;
+
+    let mut speech_samples = Vec::new();


Reserving capacity for speech_samples based on input_num_samples avoids multiple reallocations and data copies as the vector grows, improving performance.

Suggested change

let mut speech_samples = Vec::new();

let mut speech_samples = Vec::with_capacity(input_num_samples as usize);

gemini-code-assist · 2026-07-21T16:41:39Z

+if [ ! -f "./ten-vad.onnx" ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/ten-vad.onnx
+fi
+
+if [ ! -f ./lei-jun-test.wav ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/lei-jun-test.wav
+fi


By default, curl does not return a non-zero exit code on HTTP errors (like 404 or 500) unless -f or --fail is specified. Adding -f ensures the script fails immediately if the download fails, preventing subsequent commands from trying to use invalid/empty files.

Suggested change

if [ ! -f "./ten-vad.onnx" ]; then

curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/ten-vad.onnx

fi

if [ ! -f ./lei-jun-test.wav ]; then

curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/lei-jun-test.wav

fi

if [ ! -f "./ten-vad.onnx" ]; then

curl -SLf -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/ten-vad.onnx

fi

if [ ! -f ./lei-jun-test.wav ]; then

curl -SLf -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/lei-jun-test.wav

fi

… errors

csukuangfj

Thank you for your contribution!

Add ten-vad Rust API example to remove silences from a file

4a28f0c

dosubot Bot added the size:L This PR changes 100-499 lines, ignoring generated files. label Jul 21, 2026

sportiz91 mentioned this pull request Jul 21, 2026

[Help wanted] Add more examples for Rust API #3210

Open

14 tasks

gemini-code-assist Bot reviewed Jul 21, 2026

View reviewed changes

Validate 16 kHz input, reserve output capacity, and fail curl on HTTP…

beffc47

… errors

sportiz91 mentioned this pull request Jul 21, 2026

Rust crate: prebuilt static libs require GLIBCXX_3.4.29+ (GCC 11+) — document or detect at build time? #3779

Open

csukuangfj approved these changes Jul 23, 2026

View reviewed changes

csukuangfj merged commit 64cc667 into k2-fsa:master Jul 23, 2026
1 check passed

This was referenced Aug 12, 2026

Bump org.k2fsa.sherpa.onnx from 1.13.2 to 1.13.5 AIRsLight/VRChatVoiceInput#9

Closed

deps: Bump org.k2fsa.sherpa.onnx from 1.13.4 to 1.13.5 kellylford/LiveCaptionsWithAccessibility#9

Closed

dependabot Bot mentioned this pull request Aug 21, 2026

Bump org.k2fsa.sherpa.onnx from 1.13.4 to 1.13.5 shawnzhang-lab/VoiceMemo-X#8

Open

This was referenced Sep 15, 2026

deps: Bump org.k2fsa.sherpa.onnx from 1.13.4 to 1.13.8 kellylford/LiveCaptionsWithAccessibility#13

Open

Bump org.k2fsa.sherpa.onnx from 1.13.2 to 1.13.8 AIRsLight/VRChatVoiceInput#12

Open

Navigation Menu

Add ten-vad Rust API example to remove silences from a file - #3778

Add ten-vad Rust API example to remove silences from a file#3778
csukuangfj merged 2 commits into
k2-fsa:masterfrom
sportiz91:rust-ten-vad-example

sportiz91 commented Jul 21, 2026 •

edited by coderabbitai Bot

Loading

Uh oh!

coderabbitai Bot commented Jul 21, 2026 •

edited

Loading

Walkthrough

Changes

Sequence Diagram(s)

Uh oh!

gemini-code-assist Bot left a comment

Uh oh!

gemini-code-assist Bot Jul 21, 2026

Uh oh!

gemini-code-assist Bot Jul 21, 2026

Uh oh!

gemini-code-assist Bot Jul 21, 2026

Uh oh!

csukuangfj left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

	let mut speech_samples = Vec::new();
	let mut speech_samples = Vec::with_capacity(input_num_samples as usize);

Conversation

sportiz91 commented Jul 21, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary by CodeRabbit

Uh oh!

coderabbitai Bot commented Jul 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

Sequence Diagram(s)

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

gemini-code-assist Bot Jul 21, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot Jul 21, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot Jul 21, 2026

Choose a reason for hiding this comment

Uh oh!

csukuangfj left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

sportiz91 commented Jul 21, 2026 •

edited by coderabbitai Bot

Loading

coderabbitai Bot commented Jul 21, 2026 •

edited

Loading