Plugins
Parcourez, filtrez et installez des plugins DeepSeek-Harness.
211 plugins trouvés
deepseek-moonpet
femad-6/deepseek-moonpet
MoonPet (月伴), compagnon d’état et minuteur Pomodoro pour DeepSeek Harness : expressions liées à l’état, suivi du temps de travail, sous-titres multilingues et rappels de pause.
dsh-gptsovits-tts
nuoyi207/dsh-gptsovits-tts
Synthèse vocale GPT-SoVITS locale pour DSH : lit les réponses à voix haute, permet de changer de voix dans la barre latérale droite et d’en entraîner une nouvelle à partir d’un échantillon audio. Le premier lancement télécharge un moteur de 11,4 Go et nécessite un GPU NVIDIA CUDA.
dsh-voice-doubao
fyzb2020/dsh-voice-doubao
Ajoute deux services de reconnaissance vocale cloud Volcengine Doubao au sélecteur de saisie vocale de DeepSeek Harness, avec réécriture facultative de la transcription par un LLM.
dsh-stt
try-works/dsh-stt
Transcription vocale locale multi-modèles pour le compositeur DSH Web. L'utilisateur choisit Whistle ou Voz ; chaque modèle est téléchargé à la demande. Fourni en bundle DSH.
dsh-multi-tts
wyr-233/dsh-multi-tts
Lecture vocale de chaque réponse avec page de réglages multi-fournisseur : MiniMax ou tout endpoint `/audio/speech` compatible OpenAI, choix de voix, émotion, vitesse et modèle, plus option de lecture automatique.
dsh-voice-alert
loyalchiiina/dsh-voice-alert
Joue une voix/un son à la fin d’un tour et un signal d’échec si un appel d’outil ou le tour échoue ; les interruptions utilisateur restent silencieuses et les sessions subagent sont ignorées par défaut. Trois modes : voice (Volcengine TTS, voix clonées ou prédéfinies, fingerprint réutilisé pour éviter de resynthétiser un texte inchangé), effects (20 sons intégrés assignables à done/fail/approval, aperçu par ligne, sans API key ; approval n’a pas de déclencheur automatique et ne joue que manuellement) et off. Lecture par winmm sans modifier le volume système. Windows uniquement ; kernel DSH 0.2.2-rc.1.
dsh-voice-talk
duoduoqian708/dsh-voice-talk
Conversation vocale pour l’interface Web DSH : touchez le micro pour lancer un appel plein écran, parlez mains libres et écoutez chaque réponse au fil de sa génération. Interface chinois/anglais, thèmes clair/sombre, débit réglable et voix sélectionnables.
dsh-fish-tts
mari23333/dsh-fish-tts
Lit à voix haute les réponses de l’assistant uniquement via l’API Fish Audio (clé à fournir soi-même) : lecture par message, bascule de lecture automatique et page de réglages pour le modèle, le reference_id de la voix, la clé API chiffrée et le proxy.
dsh-whale-live
lh1314-china/dsh-whale-live
Left-sidebar live console for DeepSeek Harness: stream log, Bilibili danmaku, edge-tts speech, optional VTube Studio/OBS/livehime helpers and a memory inbox. Windows only.
dsh-bibibobo
cxv117793-blip/dsh-bibibobo
Speaks DSH replies aloud with the bibibobo synthesizer: one sawtooth wave and two band-pass filters, no samples and no TTS model. The control sits in the answer's own action row beside Copy, with a progress bar and a scrolling subtitle while it plays. Each turn keeps its own control, but it reads only that turn's closing reply, so a multi-step answer never restarts mid-sentence and intermediate steps get no control. Deterministic by design: a character always yields the same pitch and mouth shape. Markdown is stripped before speaking, replies are capped at 1200 characters, and playback can be paused, resumed or stopped. Zero dependencies, MIT, no network access.
dsh-voice
phoenixym0307/dsh-voice
Adds a mic button to the chat composer that turns speech into text in the input box, and optionally reads each assistant reply aloud. Speech recognition and synthesis call DeepSeek's official ASR and TTS endpoints, with the API key entered in the plugin settings page. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-go-fish-panel
27synthesis/dsh-go-fish-panel
Native panel for the dsh-fish companion on DSH Go: one observatory page showing what she has been up to, plus rhythm, browsing, and speech-and-memory settings pages that write 57 config keys in place.
smart-dictation-dsh (smart-dictation-bundle)
triagmas/smart-dictation-dsh
Local Russian dictation for the DSH composer with GigaAM v3, English technical terms, punctuation, and a personal glossary that learns only after confirmation. Audio and transcripts never leave the machine.
dsh-live-speak
chaowei08/dsh-live-speak
Reads assistant replies aloud while they stream: sentence-by-sentence speech with highlighted lines in a sidebar player, pause, mute and stop controls, and automatic quieting for code blocks. The speech engine is switchable: Edge TTS online by default, or the operating system local voices. Its HTTP routes do not verify the caller, so any program that can reach the local DSH port can call them.
dsh-breeze-speak
rirtue/dsh-breeze-speak
Reads each reply aloud through the BreezeBlue / Breeze TTS 2 API. Synthesis starts as soon as the reply first sentence is complete rather than after the whole reply, and the audio is streamed as raw PCM into Web Audio so it plays continuously while it arrives. Needs a BreezeBlue API key, entered on the settings page or via BREEZE_API_KEY; a key saved on the settings page is stored as plain text in ~/.dsh/breeze-speak.json. By default only the first sentence of each reply (up to 120 characters) is spoken, and that text is sent to api.breeze.blue. The default voice is just an example, so pick one available to your account on the settings page.
dsh-gigaam-stt
oldschool686/dsh-gigaam-stt
Local Russian speech-to-text for DeepSeek Harness voice input: GigaAM v3 (ONNX int8) on CPU via sherpa-onnx
dsh-minimax-voice
sunshine334419520/dsh-minimax-voice
MiniMax voice studio for DeepSeek Harness: clone a voice from an audio sample, preview it from the audio returned by the clone request itself (no extra synthesis call, so the per-voice fee stays uncharged), synthesize text to local mp3 files, transcribe audio to text with asr-1.0, and manage the voice library, API key and these actions from a settings panel; four agent tools and a local HTTP API.
pocket-harness (pocket-harness)
regrevia/pocket-harness
Pocket Harness: the phone side of DSH. Continue the same DSH session from your phone, with the same history, projects, goal, queue and pending questions, connected over your LAN, your own domain or a relay you run (no author-run server, no account). Needs the Android companion app (APK from the repo's releases) paired by QR code; an optional pixel-whale desktop pet can talk and listen, and neural voice and local speech recognition are optional extras. Formerly dsh-pet-voice.
dsh-shortvideo-export
hangyesite/dsh-shortvideo-export
Download watermark-free Douyin and Xiaohongshu videos, transcribe them with local or cloud speech recognition, proofread the text against a term list, and export Markdown, DOCX, SRT, VTT and other formats from a settings panel or agent tools. Needs Python 3.10+ with pip packages (playwright, python-docx, numpy, and sherpa-onnx for the default local SenseVoice engine, which also downloads a ~228 MB model on first use), ffmpeg on PATH, and Chrome or Edge.
dsh-doubao-video
fangqiuci/dsh-doubao-video
Doubao video workshop: a gated pipeline from storyboard to finished cut. Locked art direction, shot validation, character-card vs reference-image separation, first/last-frame continuity and clip dedup; model-aware duration caps (Mini/Fast 10s, 2.0 15s, 2.5 40s). Uses ffmpeg for frame extraction, concatenation and subtitles; speech synthesis calls a separately installed local TTS plugin, and ASR subtitles need Python with faster-whisper.
ai-voice
root-7-khx/ai-voice
A local AI voice studio for DeepSeek Harness with nine local TTS engines for voice design, voice cloning, emotional performance, a reference-audio library and audio export. It only registers /ai-voice/api routes; the UI lives in the companion-memory plugin's integrations page, and every engine needs its own Python virtualenv and downloaded model weights, which are not bundled.
dsh-live2d-stage
ikunua/dsh-live2d-stage
Renders a Live2D Cubism 4 model as the DSH web UI background: a transparent click-through canvas that covers the window without blocking any control. Eyes, head and body follow the pointer; expressions track the agent state (thinking, running a tool, turn finished, error, waiting) and the conversation (message arrives, reply ends, question asked); after an idle delay the character shows speech bubbles and plays motions and expressions drawn at random from pools you tick. Settings list the models under ~/.dsh/live2d-stage/models, expose position, scale, opacity, gaze strength and every interval and probability, and open a resizable live preview of the same canvas. Expressions and motions a model3.json omits are detected from the model folder; one sample model is included.
dsh-jianxian-widget (plugin)
cococo1ng/dsh-jianxian-widget
Qingyun Jianxian (青云剑仙, Azure Cloud Sword Immortal) — a resident bottom-right widget for the DeepSeek Harness Web UI: a sword-cultivator character that reads your account balance and token spend, keeps a local ledger broken down by model, and has 10 switchable sound effects. Reminders cover low balance, a model waiting on your answer, a pending approval, and staying up too late. Ships 23 idle poses plus a 24-frame sword dance, 22 menu-icon skins, six panel tabs, a line-by-line speech-bubble editor, and black or white robes that follow the light or dark theme. Runs entirely locally and registers no agent tools.
dsh-go-local-voice
27synthesis/dsh-go-local-voice
Synthèse vocale pour l’interface mobile native dsh-go : un appui long sur une réponse de l’assistant la lit avec Xiaomi MiMo TTS ou ByteDance Doubao Seed TTS. Après chaque tour, la réponse est segmentée et pré-synthétisée en arrière-plan ; un seul appui lance un MP3 continu. Un lease de lecture stable permet l’arrêt, une bande d’état et une barre de progression affichent l’avancement, et l’audio peut être exporté sur le téléphone. Les clés restent sur l’appareil en mode 0600, ne sont ni affichées, ni commitées, ni envoyées au modèle.