Update DPDFNet documentation

[danielr-ceva]

Summary

Update the DPDFNet documentation to reflect the changes introduced in
k2-fsa/sherpa-onnx #3824.

The documentation now:

• Covers the new 8 kHz, 16 kHz, and 48 kHz DPDFNet model variants.
• Separates the model variants into tables by sample rate.
• Updates model download instructions and examples.
• Documents the offline attenuation-limit option.
• Updates the Python, C, JavaScript, and Rust examples.

Summary by CodeRabbit

• Documentation
  • Expanded DPDFNet speech enhancement guidance for 8, 16, and 48 kHz models.
  • Documented offline attenuation limits, defaults, valid values, and configuration options.
  • Added streaming enhancement examples and clarified resampling and sample-rate requirements.
  • Updated C, Python, JavaScript, and Rust examples with current configuration and behavior.
  • Added official model download guidance and links to supporting resources.

[coderabbitai]

Note

Currently processing new changes in this PR. This may take a few minutes, please wait...

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 5eb96f8d-a11c-43dd-a2e4-653f5203b4ea

📥 Commits

Reviewing files that changed from the base of the PR and between 4a1db71 and dbc624b.

📒 Files selected for processing (7)

• docs/source/onnx/javascript-api/examples/api_speech_denoiser.rst
• docs/source/onnx/javascript-api/examples/speech_enhancement.rst
• docs/source/onnx/rust-api/examples/offline_speech_enhancement_dpdfnet.rst
• docs/source/onnx/rust-api/examples/streaming_speech_enhancement_dpdfnet.rst
• docs/source/onnx/speech-enhancement/dpdfnet-c-api.rst
• docs/source/onnx/speech-enhancement/dpdfnet-python-api.rst
• docs/source/onnx/speech-enhancement/dpdfnet.rst

 _____________________________________________
< Nose to the grindstone, eyes on the screen. >
 ---------------------------------------------
  \
   \   (\__/)
       (•ㅅ•)
       / 　 づ

✨ 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

• X
• Mastodon
• Reddit
• LinkedIn

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

[csukuangfj]

Thank you for your contribution!

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@docs/source/onnx/speech-enhancement/dpdfnet.rst`:
- Around line 115-117: Update the download commands in the DPDFNet documentation
to reference valid, published model asset URLs. Verify all three 8 kHz and 48
kHz links resolve successfully before retaining them in the instructions.

🪄 Autofix

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 5eb96f8d-a11c-43dd-a2e4-653f5203b4ea

📥 Commits

Reviewing files that changed from the base of the PR and between 4a1db71 and dbc624b.

📒 Files selected for processing (7)

• docs/source/onnx/javascript-api/examples/api_speech_denoiser.rst
• docs/source/onnx/javascript-api/examples/speech_enhancement.rst
• docs/source/onnx/rust-api/examples/offline_speech_enhancement_dpdfnet.rst
• docs/source/onnx/rust-api/examples/streaming_speech_enhancement_dpdfnet.rst
• docs/source/onnx/speech-enhancement/dpdfnet-c-api.rst
• docs/source/onnx/speech-enhancement/dpdfnet-python-api.rst
• docs/source/onnx/speech-enhancement/dpdfnet.rst

[coderabbitai]

🎯 Functional Correctness | 🟠 Major | ⚡ Quick win

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
set -eu

base="https://github.com/k2-fsa/sherpa-onnx/releases/download/speech-enhancement-models"
for model in \
  dpdfnet2_8khz.onnx \
  dpdfnet8_8khz.onnx \
  dpdfnet8_48khz_hr.onnx
do
  status="$(curl -L -s -o /dev/null -w '%{http_code}' "$base/$model")"
  test "$status" = 200 || {
    echo "$model: HTTP $status"
    exit 1
  }
done

Repository: k2-fsa/sherpa

Length of output: 180

---

🏁 Script executed:

#!/bin/bash
base="https://github.com/k2-fsa/sherpa-onnx/releases/download/speech-enhancement-models"
for model in dpdfnet2_8khz.onnx dpdfnet8_8khz.onnx dpdfnet8_48khz_hr.onnx; do
  status="$(curl -L -s -o /dev/null -w '%{http_code}' "$base/$model")"
  printf '%s: HTTP %s\n' "$model" "$status"
done

Repository: k2-fsa/sherpa

Length of output: 242

---

Publish or replace the new model assets before merging.

All three release URLs return HTTP 404. The download instructions are unusable.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@docs/source/onnx/speech-enhancement/dpdfnet.rst` around lines 115 - 117,
Update the download commands in the DPDFNet documentation to reference valid,
published model asset URLs. Verify all three 8 kHz and 48 kHz links resolve
successfully before retaining them in the instructions.

[danielr-ceva]

Thanks @csukuangfj for the quick merge! :)

Just to make sure this doesn’t fall through the cracks - will you also download the new/updated models and add them to the repo?

The models are:

• dpdfnet2_8khz.onnx - new
• dpdfnet2_48khz_hr.onnx - updated
• dpdfnet8_8khz.onnx - new
• dpdfnet8_48khz_hr.onnx - new