Parlova runs on your own machine — the client starts it, so there is no endpoint to ping. 688 installs a week from pypi. Last commit 11 Nov 2026.
Transcribe local audio/video and YouTube into timestamped, speaker-labelled text. 90+ languages.
Today is the operative word: we check Parlova every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add parlova-mcp -- uvx parlova
{
"mcpServers": {
"parlova-mcp": {
"command": "uvx",
"args": [
"parlova"
]
}
}
}
[mcp_servers.parlova-mcp]
command = "uvx"
args = ["parlova"]
{
"mcpServers": {
"parlova-mcp": {
"command": "uvx",
"args": [
"parlova"
]
}
}
}
{
"servers": {
"parlova-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"parlova"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
PARLOVA_TOKEN (Access token from Parlova > Settings > Developers (Plus or Pro plan). Optional if you ran `parlova login`).
Transcribe public videos & audio (YouTube, TikTok, IG) into accurate, timestamped text via API.
Turn any YouTube video into text: transcript, title, channel, timestamps.
Search and cite timestamped YouTube and podcast transcripts; transcribe any video or episode.
YouTube transcripts for AI agents: text and timestamps for any video, channel or playlist.
YouTube video and Shorts transcripts with timestamps. Paid, on Apify's hosted MCP.
Search YouTube, get video metadata, and extract timestamped transcripts for AI workflows.
YouTube transcripts for AI agents: captions with timestamps, languages, TXT, SRT and VTT export.
Transcripts of YouTube videos, playlists and channels in any language: text, SRT, VTT or JSON.
Answers built from our own checks of this server.