ArchivedThis repository is archived and may no longer be actively maintained.
← DeepSeek Harness Plugins
DeepSeek Harness PluginManifest Valid

kepos-speech

Speech synthesis and short-audio recognition plugin for DeepSeek Harness Web using Alibaba and ByteDance services.

UI & ProductivityBrowser & WebRemote Execution
0GitHub Stars0ForksUpdated2026-09-04

Install

$ dsh plugin --profile <profile> add @lamplitisles/kepos-speech

Plugin Overview & Capabilities

AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.

source-grounded
Kepos Speech provides Chinese speech synthesis (TTS) and short-audio recognition (ASR) capabilities within DeepSeek Harness Web. It processes TTS marker blocks in assistant responses to render inline audio players, provides bounded Qwen ASR transcription for user audio attachments, and exposes an optional Cordis host service for in-process speech tasks.

Key Capabilities

  • Synthesizes Chinese speech from tagged assistant response blocks using Alibaba or ByteDance TTS
  • Transcribes short user audio attachments with expression and language metadata using Qwen ASR
  • Provides in-browser playback UI integrated with DeepSeek Harness Web
  • Caches prepared MP3 audio files under the active session workspace
  • Exposes an optional in-process Cordis service (keposSpeech) for Host synthesis and transcription calls

Useful For

  • Listening to assistant responses via synthesized voice in Chinese
  • Transcribing voice audio messages directly into conversational turns
  • Accessing programmatic speech synthesis and transcription services from other DSH plugins

Who It Fits

  • DeepSeek Harness Web users wanting voice interaction and playback
  • Developers building voice-enabled workflows and extensions inside DSH

Documented Limitations

  • Repository is archived and may not receive further updates
  • Requires active DashScope (Alibaba) or Volcengine (ByteDance) API credentials
  • ASR recognition audio is strictly bounded up to 10 MB per attachment
  • TTS provider options are limited to Alibaba DashScope and ByteDance Volcengine

DSH Compatibility

Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.

Not tested yetNo runtime compatibility tests have been published yet.

Security Signals

Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.

Dependency Counts

package.json declares 0 runtime, 12 development, 19 peer, and 0 optional dependencies.

info
Dsh Bundle Declared

package.json declares DSH bundle metadata.

info
License Declared

GitHub reports the repository license as Apache-2.0.

info
Package Manifest Available

A root package.json was captured and can be inspected by the registry.

info
Repository Archived

GitHub reports this repository as archived.

medium
Source Available

Public GitHub source metadata is available for this registry snapshot.

info

Source & Registry Notes

Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.

Source repository
LamplitIsles/kepos-speech
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin
Source checked
06f5960 · 2026-09-12

This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.

Repository Activity

GitHub stars
0GitHub stars
Forks
0
Open issues
0
Last commit
2026-09-04
Last release
No release detected
For maintainers

Maintaining this plugin?

This listing is generated from public repository data. If you maintain this project, you can review the information and share this listing with your users if you find it useful.

Add to README
Listed on DSHPlugin.app