Sherpa Onnx Tts Andy27725

Local text-to-speech via sherpa-onnx (offline, no cloud)

Security Scan

The skill's requirements and instructions are generally consistent with an offline sherpa-onnx TTS wrapper, but there are a few implementation/packaging inconsistencies you should be aware of before installing.

Purpose & Capability

The name/description (local/offline TTS via sherpa-onnx) match the declared env vars (SHERPA_ONNX_RUNTIME_DIR, SHERPA_ONNX_MODEL_DIR) and the install steps (download runtime and model archives from the project's GitHub releases). Nothing requested appears unrelated to running an offline TTS.

Instruction Scope

SKILL.md stays within TTS setup and usage (downloading runtime/models, setting env vars, running the CLI). However, it states 'The wrapper lives in this skill folder' and references {baseDir}/bin/sherpa-onnx-tts, yet no code files or wrapper are included in the package — an implementation gap. The notes mention optional env variables (e.g., SHERPA_ONNX_MODEL_FILE) that are not declared in requires.env but are optional runtime knobs.

Install Mechanism

Install steps are download-and-extract of release archives hosted on GitHub (k2-fsa/sherpa-onnx). Using GitHub releases is a standard approach; archives will be written to disk and extracted (extract: true), which is expected for this task. No obscure third-party or shortened URLs were used.

Credentials

The two required env vars correspond directly to the runtime and model locations the skill needs. No unrelated secrets or credentials are requested. A few optional env names appear in the docs but are not required.

Persistence & Privilege

always is false and the skill does not request special persistent privileges or system-wide configuration changes beyond advising edits to the user's ~/.openclaw/openclaw.json. Nothing attempts to modify other skills or system auth.

Assessment

This skill appears to be a local/offline TTS wrapper that downloads the sherpa-onnx runtime and a voice model from the project's GitHub releases — which is expected. Before installing: (1) note that the SKILL.md claims a wrapper script/binary in the skill folder but the package contains no code files; you may need to run the runtime CLI directly or supply your own wrapper. (2) Verify the GitHub release URLs and consider checking checksums or release signatures if available before running extracted binaries. (3) Understand that archives will be downloaded and extracted into your ~/.openclaw/tools/... directory and you will set environment variables pointing to them. (4) If you require a different voice, pick another model from the same release; optional env vars mentioned in the docs are not required but can be used at runtime. If you want stronger assurance, ask the publisher for the missing wrapper or a package that includes the wrapper script, and/or request release checksums or a signed release.

sherpa-onnx-tts

Local TTS using the sherpa-onnx offline CLI.

Install

1. Download the runtime for your OS (extracts into ~/.openclaw/tools/sherpa-onnx-tts/runtime)
2. Download a voice model (extracts into ~/.openclaw/tools/sherpa-onnx-tts/models)

Update ~/.openclaw/openclaw.json:

{
  skills: {
    entries: {
      "sherpa-onnx-tts": {
        env: {
          SHERPA_ONNX_RUNTIME_DIR: "~/.openclaw/tools/sherpa-onnx-tts/runtime",
          SHERPA_ONNX_MODEL_DIR: "~/.openclaw/tools/sherpa-onnx-tts/models/vits-piper-en_US-lessac-high",
        },
      },
    },
  },
}

The wrapper lives in this skill folder. Run it directly, or add the wrapper to PATH:

export PATH="{baseDir}/bin:$PATH"

Usage

{baseDir}/bin/sherpa-onnx-tts -o ./tts.wav "Hello from local TTS."

Notes:

• Pick a different model from the sherpa-onnx tts-models release if you want another voice.
• If the model dir has multiple .onnx files, set SHERPA_ONNX_MODEL_FILE or pass --model-file.
• You can also pass --tokens-file or --data-dir to override the defaults.
• Windows: run node {baseDir}\\bin\\sherpa-onnx-tts -o tts.wav "Hello from local TTS."