How do I use mcp-gladia?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-gladia Transcribe my latest podcast episode and summarize it." That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

mcp-gladia

Official

by gladiaio

Overview Schema Source Related Servers Discussions Admin

TypeScript

Local

mcp-gladia

MCP server for Gladia audio transcription and intelligence. Enables LLMs to transcribe, analyze, and translate audio/video content through Gladia's API.

Getting Your API Key

Sign up at app.gladia.io
Navigate to the API Keys section
A default API key is automatically created for new accounts

Gladia offers 10 hours of free audio transcription per month. No credit card required.

Related MCP server: Deepgram MCP Server

Installation

npm install mcp-gladia

Or run directly:

npx mcp-gladia

Configuration

Set your Gladia API key as an environment variable:

export GLADIA_API_KEY=your-api-key-here

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "gladia": {
      "command": "npx",
      "args": ["mcp-gladia"],
      "env": {
        "GLADIA_API_KEY": "your-api-key-here"
      }
    }
  }
}

Claude Code

claude mcp add gladia -- npx mcp-gladia

Then set your API key in the environment or .env file.

Other MCP Clients

Any MCP-compatible client can use this server via stdio transport. Set the command to npx mcp-gladia and provide GLADIA_API_KEY as an environment variable.

Tools

upload_file

Upload an audio/video file to Gladia for transcription.

Parameter	Type	Required	Description
filePath	string	Yes	Path to the audio/video file

Supported formats: mp3, wav, m4a, mp4, mov, avi, flac (max 1GB).

transcribe

Submit an audio file for transcription with automatic polling until completion. Supports all Gladia audio intelligence features. Returns the completed result or a job ID if timeout occurs (5 min).

Parameter	Type	Required	Description
audioUrl	string	Yes	URL from upload_file
language	string	No	Language code (e.g. `en`, `fr`). See supported languages
detectLanguage	boolean	No	Auto-detect language (default: true)
diarization	boolean	No	Enable speaker identification
diarizationConfig	object	No	`{ numberOfSpeakers?, minSpeakers?, maxSpeakers? }`
subtitles	boolean	No	Generate subtitle files
subtitlesConfig	object	No	`{ formats: ["srt", "vtt"] }`
customVocabulary	string[]	No	Custom words to improve recognition
summarization	boolean	No	Enable transcription summary
summarizationConfig	object	No	`{ type: "general" \| "concise" \| "bullet_points" }`
sentimentAnalysis	boolean	No	Enable sentiment/emotion analysis
namedEntityRecognition	boolean	No	Enable entity detection
chapterization	boolean	No	Enable chapter detection with timestamps
translation	boolean	No	Enable translation
translationConfig	object	No	`{ targetLanguages: ["fr", "es"], model?: "base" \| "enhanced" }`
audioToLlm	boolean	No	Enable custom LLM analysis
audioToLlmConfig	object	No	`{ prompts: ["your question about the audio"] }`

get_transcription_status

Check the status of a transcription job (useful for long-running jobs that timed out).

Parameter	Type	Required	Description
jobId	string (UUID)	Yes	Job ID from a previous transcribe request

list_transcription_jobs

List past transcription jobs with optional filtering.

Parameter	Type	Required	Description
offset	number	No	Pagination offset
limit	number	No	Max results (default: 20)
status	string	No	Filter: `queued`, `processing`, `done`, `error`
afterDate	string	No	Filter by creation date (ISO 8601)
beforeDate	string	No	Filter by creation date (ISO 8601)
kind	string	No	Filter: `pre-recorded`, `live`

delete_transcription_job

Delete a transcription job and its data.

Parameter	Type	Required	Description
jobId	string (UUID)	Yes	Job ID to delete

Audio Intelligence Features

All intelligence features are enabled as options on the transcribe tool and processed server-side by Gladia.

Summarization

Generate a summary of the transcription in one of three formats:

Type	Description
`general`	Balanced, comprehensive summary (default)
`concise`	Short overview of key points
`bullet_points`	Key takeaways as a bullet list

{ "summarization": true, "summarizationConfig": { "type": "bullet_points" } }

Sentiment & Emotion Analysis

Detect sentiment and emotion for each sentence in the transcript, with speaker attribution when diarization is enabled.

Sentiments: positive, negative, neutral, mixed, unknown

Emotions: adoration, anger, joy, fear, surprise, sadness, neutral, and more

{ "sentimentAnalysis": true }

Named Entity Recognition

Detect and classify entities mentioned in the audio. Supports 50+ entity types across multiple categories:

Category	Entity Types
PII	Name, Email, Phone Number, SSN
Location	City, Country, Address
Medical (PHI)	Conditions, Drugs, Injuries
Financial	Bank Account, Credit Card
Demographic	Age, Gender, Occupation
Temporal	Date, Time

Supports GDPR, HIPAA, and CPRA compliance workflows.

{ "namedEntityRecognition": true }

Chapterization

Automatically segment audio into logical chapters. Each chapter includes:

Summary — overview of the chapter content
Headline — short title
Gist — one-line bottom line
Keywords — key terms mentioned
Timestamps — start and end times

{ "chapterization": true }

Translation

Translate transcriptions to one or more target languages.

Model	Description
`base`	Fast translation, covers most use cases
`enhanced`	Higher quality, better for complex content

{ "translation": true, "translationConfig": { "targetLanguages": ["fr", "es"], "model": "enhanced" } }

Audio-to-LLM (Custom Prompts)

Run custom analysis prompts directly against the audio content. No need to post-process transcripts with a separate LLM.

{
  "audioToLlm": true,
  "audioToLlmConfig": {
    "prompts": [
      "Extract the key decisions made in this meeting",
      "What are the action items and who is responsible?"
    ]
  }
}

Speaker Diarization

Identify and separate speakers in the audio. Output includes speaker labels on every utterance.

{
  "diarization": true,
  "diarizationConfig": { "minSpeakers": 2, "maxSpeakers": 5 }
}

Supported Languages

100+ languages supported for transcription. Use the language code with the language parameter, or set detectLanguage: true (default) for automatic detection.

Language	Code	Language	Code	Language	Code
Afrikaans	`af`	Hawaiian	`haw`	Persian	`fa`
Albanian	`sq`	Hebrew	`he`	Polish	`pl`
Amharic	`am`	Hindi	`hi`	Portuguese	`pt`
Arabic	`ar`	Hungarian	`hu`	Punjabi	`pa`
Armenian	`hy`	Icelandic	`is`	Romanian	`ro`
Assamese	`as`	Indonesian	`id`	Russian	`ru`
Azerbaijani	`az`	Italian	`it`	Sanskrit	`sa`
Bashkir	`ba`	Japanese	`ja`	Serbian	`sr`
Basque	`eu`	Javanese	`jw`	Shona	`sn`
Belarusian	`be`	Kannada	`kn`	Sindhi	`sd`
Bengali	`bn`	Kazakh	`kk`	Sinhala	`si`
Bosnian	`bs`	Khmer	`km`	Slovak	`sk`
Breton	`br`	Korean	`ko`	Slovenian	`sl`
Bulgarian	`bg`	Lao	`lo`	Somali	`so`
Catalan	`ca`	Latin	`la`	Spanish	`es`
Chinese	`zh`	Latvian	`lv`	Sundanese	`su`
Croatian	`hr`	Lingala	`ln`	Swahili	`sw`
Czech	`cs`	Lithuanian	`lt`	Swedish	`sv`
Danish	`da`	Luxembourgish	`lb`	Tagalog	`tl`
Dutch	`nl`	Macedonian	`mk`	Tajik	`tg`
English	`en`	Malagasy	`mg`	Tamil	`ta`
Estonian	`et`	Malay	`ms`	Tatar	`tt`
Faroese	`fo`	Malayalam	`ml`	Telugu	`te`
Finnish	`fi`	Maltese	`mt`	Thai	`th`
French	`fr`	Maori	`mi`	Tibetan	`bo`
Galician	`gl`	Marathi	`mr`	Turkish	`tr`
Georgian	`ka`	Mongolian	`mn`	Turkmen	`tk`
German	`de`	Myanmar	`my`	Ukrainian	`uk`
Greek	`el`	Nepali	`ne`	Urdu	`ur`
Gujarati	`gu`	Norwegian	`no`	Uzbek	`uz`
Haitian Creole	`ht`	Nynorsk	`nn`	Vietnamese	`vi`
Hausa	`ha`	Occitan	`oc`	Welsh	`cy`
		Pashto	`ps`	Wolof	`wo`
				Yiddish	`yi`
				Yoruba	`yo`

Usage Examples

Basic Transcription

Upload my-recording.mp3 and transcribe it

Meeting with Multiple Speakers

Transcribe this meeting recording with diarization enabled, expecting 3-5 speakers.
Generate a bullet-point summary and extract action items using audio-to-LLM.

Multilingual Content Analysis

Transcribe this podcast, detect the language, translate to English and French,
and run sentiment analysis on the conversation.

Compliance & Entity Detection

Transcribe this customer call with named entity recognition to identify
any PII mentioned (names, emails, phone numbers).

Custom Audio Analysis

Transcribe this earnings call and use audio-to-LLM with these prompts:
- "What are the key financial metrics mentioned?"
- "What is the company's guidance for next quarter?"
- "Summarize the Q&A section"

Troubleshooting

Issue	Solution
`GLADIA_API_KEY is required`	Set the `GLADIA_API_KEY` environment variable
`Unsupported file format`	Use mp3, wav, m4a, mp4, mov, avi, or flac
`File too large`	Files must be under 1GB
Transcription timeout	Use `get_transcription_status` with the returned job ID
Translation fails	Ensure `translationConfig.targetLanguages` is provided
`Invalid uuid`	Job IDs must be valid UUIDs (from transcribe or list_transcription_jobs)

Development

git clone https://github.com/gladiaio/mcp-gladia.git
cd mcp-gladia
npm install
npm run build
npm run dev

Requires Node.js 18+.

License

MIT

This server cannot be deployed

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Maintenance

ActivityInactive

ResponsivenessSyncing

Resources

GitHub Repository

Need Help?

Related Servers

Related MCP Connectors

transcribeOAuth
com.transcribevideototext
Transcribe audio & video to text for AI agents: 100+ languages, speaker labels, webhooks.
Speech Processing Audio Processing Multimedia Processing
2
mcp A
online.editclips
Video, audio, and image processing for AI agents: convert, transcribe, upscale - 150+ operations.
Multimedia Processing Audio Processing Image & Video Processing
51
mcpOAuth
so.transcribe
Transcribe audio and video into speaker-labelled transcripts, subtitles, clips, and cited Q&A.
Speech Processing Audio Processing
1
justtranscribe-mcpOAuth
io.github.bramontiventures
Transcribe public videos & audio (YouTube, TikTok, IG) into accurate, timestamped text via API.
Speech Processing Audio Processing
1

Related MCP Servers

MCP-Transcribe
Speech Processing Audio Processing
transcribe-app
A
license
Not graded
quality
C
maintenance
This service provides fast and reliable transcriptions for audio/video files and voice memos. It allows LLMs to interact with the text content of audio/video file.
Updated 5 months ago (2026-04-03 12:27 UTC)
8
MIT
Deepgram MCP Server
Speech Processing Audio Processing Multimedia Processing
reddheeraj
A
license
Not graded
quality
D
maintenance
Enables speech-to-text transcription, text-to-speech synthesis, and audio analysis using Deepgram's AI models. Supports features like speaker diarization, sentiment analysis, language detection, and various audio processing capabilities.
Updated a year ago (2025-09-16 18:39 UTC)
2
MIT
IRCAM Amplify MCP Server
Audio Processing Multimedia Processing
pixxelboy
A
license
A
quality
Not graded
maintenance
Enables LLMs to analyze music (genre, mood, tempo, key), separate audio stems, detect AI-generated music, and measure loudness using IRCAM Amplify's audio processing APIs.
Updated 9 months ago (2025-12-13 09:12 UTC)
5
-
@vocea.app/mcp-serverofficial
Speech Processing Text-to-Speech
vocea-admin
A
license
A
quality
C
maintenance
Enables AI agents to generate speech, transcribe audio, and manage voices via the Vocea API.
Updated 4 months ago (2026-05-14 18:20 UTC)
6
MIT

mcp-gladia

Getting Your API Key

Installation

Configuration

Claude Desktop

Claude Code

Other MCP Clients

Tools

upload_file

transcribe

get_transcription_status

list_transcription_jobs

delete_transcription_job

Audio Intelligence Features

Summarization

Sentiment & Emotion Analysis

Named Entity Recognition

Chapterization

Translation

Audio-to-LLM (Custom Prompts)

Speaker Diarization

Supported Languages

Usage Examples

Basic Transcription

Meeting with Multiple Speakers

Multilingual Content Analysis

Compliance & Entity Detection

Custom Audio Analysis

Troubleshooting

Development

Links

License

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

MCP-Transcribe

Deepgram MCP Server

IRCAM Amplify MCP Server

@vocea.app/mcp-serverofficial