about shrp
SHRP is a workspace for turning speech, audio, video, and text into useful output. It started as a simple speech-to-text tool and now includes text-to-speech, YouTube transcripts, smart voice tools, public API endpoints, and MCP workflows for AI agents.
It is built by Digiwares, a small portfolio of privacy-first tools. The goal is not to become another crowded AI hub. SHRP focuses on practical content transformation: send content in, get a useful transcript, audio file, summary, invoice, or structured result out.
// what shrp does
speech-to-text
live browser dictation and file transcription
text-to-speech
natural voices with downloadable audio
youtube to text
extract readable transcripts from video links
smart voice tools
turn speech into invoices, summaries, notes, and structured outputs
api docs
REST endpoints for transcription, TTS, and YouTube transcripts
mcp server
use SHRP from Claude Desktop, Cursor, Windsurf, and agents
// principles
fast tools first
The main tools are designed to be useful immediately: speak, paste text, upload a file, or paste a YouTube link and get output without a complicated setup.
privacy-aware by default
Microphone transcription stays in your browser. Uploaded files are processed for transcription and not kept as permanent audio storage.
one workspace
SHRP keeps related workflows together: transcripts, projects, text-to-speech, smart voice tools, API keys, and MCP access.
// contact
Questions, student/educator requests, or API feedback: hello@shrp.app