Nemotron Transcribe MCP
Nemotron Transcribe MCP lets an agent transcribe, translate, analyze, and manage jobs for audio files or YouTube links through MCP tools backed by a local/LAN vLLM Nemotron endpoint.
Transcribe audio or YouTube links with
verbatim,clean,technical, orraw_dictationstyles.Add speaker labels for 2–8 speakers and match speaker labels across long-file chunks.
Handle long recordings by splitting at pauses, processing chunks, and stitching results.
Translate speech between English and supported languages, with optional source/target language settings.
Ask questions about an audio file, optionally limited to a start/end time range, with optional reasoning output.
Identify the spoken language and list supported languages with their transcription, translation, and identification support levels.
Run long tasks in the background and check job status, retrieve paginated text results, list recent jobs, or cancel jobs.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Nemotron Transcribe MCP transcribe /data/audio/team-meeting.m4a and translate it to English"
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.
Nemotron Transcribe MCP
An MCP server for NVIDIA's Nemotron 3.5 Transcribe. It gives an agent tools for transcription, speaker labels on multi-hour recordings, translation, language identification, and questions about audio.
It talks to a vLLM endpoint you already run. Audio stays on your machine or LAN.
You: Transcribe ~/audio/team-offsite.m4a and label who's talking.
Agent: transcribe(path, speakers=2) -> job id -> get_job_status -> get_job_result
You: Between minutes 40 and 55, what did they decide about the launch date?
Agent: ask_about_audio(path, question, start=2400, end=3300)Why a server and not a script
Nemotron 3.5 Transcribe is a language model with speech and speaker encoders attached (34B parameters, about 4.3B active). One chat endpoint handles every task, and the prompt picks which one. Each task needs different prompts, sampling, and token limits, and a request can hold only about an hour of audio. This server handles that so the agent doesn't have to.
Task | Model constraint | What the server does |
Transcription | Prompt wording sets the style | Four styles from versioned prompt templates |
Speaker labels | Tags up to 8 speakers, but numbering restarts every request | Matches labels across chunks |
Long audio | About 1 hour per request | Splits at pauses into ≤55 min pieces, then joins them |
Translation | English must be one side | Checks the pair; detects the source language if not given |
Language ID | Fixed label set | Samples 10 s of speech, skipping leading silence |
Audio Q&A | Reasoning mode, 16,384-token budget | Uses the card's reasoning settings; rejects spans over ~22 min |
Related MCP server: transcriber-mcp
Architecture
flowchart LR
subgraph server["nemo-transcribe-mcp"]
Tools["Tools"] --> Jobs["Job queue (SQLite)"]
Jobs --> Planner["ffmpeg + chunking"]
Planner --> Client["Requests"]
Client --> Stitcher["Stitching + speaker matching"]
end
Agent["MCP client"] -->|"stdio or HTTP"| Tools
Client -->|"chat/completions"| Model["vLLM: Nemotron 3.5 Transcribe"]How it works
Long files. Audio is converted to 16 kHz mono PCM. Files over 55 minutes are cut at the longest pause near the end of each window. If there is no pause, the next piece starts 5 seconds early so no words are lost. Repeated words at each join are removed.
Speakers across chunks. The model tells up to eight speakers apart on its own and tags each turn in the transcript:
<spk:0> Okay, let's start with the launch date.
<spk:1> I'd push it a week. QA needs more time.
<spk:0> Agreed. Let's say the fourteenth.It numbers speakers from 0 on every request, though. With speaker labels on, each piece overlaps the previous one by 10 seconds. The overlapping words show which new label matches which old one. If the match is weak (under 60% agreement) or ambiguous, the transcript gets a [speaker-join-uncertain] marker instead of a guess.
Token limits. Audio costs 12.5 tokens per second, so max_tokens is set to whatever the 65,536-token context has left. If the output is still cut off, that span is split in half and retried.
Jobs. Long work runs in the background. Each tool waits up to wait_seconds (default 25), then returns a job id and a note telling the agent what to do next. Jobs are stored in SQLite. Repeat requests return the cached result, an interrupted job picks up where it stopped when asked again, and results can be read in pages.
Also: stdio and HTTP transports (HTTP uses a bearer token), JSON logs with secrets redacted, init and doctor setup commands, and 112 tests that run without a GPU.
Results
NVFP4 checkpoint on a DGX Spark (GB10), vLLM 0.32.0:
A 2 h 47 min podcast with speaker labels took about 18 minutes (about 9× real time).
The same sentence from a 5-minute business-news clip:
Call | Output |
| That's according to earnings calls this week at both UPS and Fed Ex, some of the biggest players in this space |
| that's according to earnings calls this week at both ups and fedex some of the biggest players in this space |
| Selon les conférences téléphoniques sur les résultats de cette semaine chez UPS et FedEx, deux des acteurs les plus importants dans ce secteur |
|
|
Tools
Tool | Inputs | Returns |
|
| Transcript or job id |
|
| Translation or job id |
|
| Answer or job id |
|
| Label such as |
| Language matrix | |
|
| Status |
|
| Result text |
| optional | Recent jobs |
|
| Status |
path can be a local file or a YouTube link. Any speakers value from 2 to 8 turns on speaker matching across pieces; the model finds the actual speakers itself.
Styles: verbatim (default) keeps fillers. clean drops fillers and repeats. technical keeps emails, URLs, paths, and IDs exact. raw_dictation has no punctuation and spells out numbers.
Setup
You need a running Nemotron 3.5 Transcribe endpoint (see the model card for vllm serve). Settings go in .env; .env.example documents each one.
Local (stdio)
Requires Python 3.12+, uv, and ffmpeg.
uv sync
cp .env.example .env # set NEMOTRON_BASE_URL
uv run nemo-transcribe-mcp doctor # checks endpoint, model, ffmpegCursor config, run from this directory:
{
"mcpServers": {
"nemo-transcribe": { "command": "uv", "args": ["run", "nemo-transcribe-mcp"] }
}
}Docker (HTTP)
cp .env.example .env # set NEMOTRON_BASE_URL and MCP_AUTH_TOKEN
docker compose up -d --build
curl -s http://127.0.0.1:8083/healthzThe MCP endpoint is http://127.0.0.1:8083/mcp with Authorization: Bearer <MCP_AUTH_TOKEN>. Put audio in ./audio and pass it as /data/audio/<file>. For a model on the same host, use host.docker.internal in NEMOTRON_BASE_URL, not localhost.
Hermes example (~/.hermes/config.yaml):
mcp_servers:
nemo-transcribe:
url: "http://127.0.0.1:8083/mcp"
headers:
Authorization: "Bearer YOUR_TOKEN_FROM_ENV"
timeout: 300Languages
The model card rates 40 languages. Norwegian is also listed below; the card tested it but didn't rate it. Any of them can be transcribed. Translation needs English on one side.
Leading
Language | Code | Translate | |
Czech | 🇨🇿 |
| ✅ |
Dutch | 🇳🇱 |
| ✅ |
English | 🇺🇸 |
| ✅ |
French | 🇫🇷 |
| ✅ |
German | 🇩🇪 |
| ✅ |
Hindi | 🇮🇳 |
| ✅ |
Italian | 🇮🇹 |
| ✅ |
Korean | 🇰🇷 |
| ✅ |
Polish | 🇵🇱 |
| ✅ |
Portuguese | 🇧🇷 |
| ✅ |
Russian | 🇷🇺 |
| ✅ |
Slovak | 🇸🇰 |
| ✅ |
Spanish | 🇪🇸 |
| ✅ |
Production-ready
Language | Code | Translate | |
Bulgarian | 🇧🇬 |
| ✅ |
Croatian | 🇭🇷 |
| ✅ |
Danish | 🇩🇰 |
| ✅ |
Estonian | 🇪🇪 |
| ✅ |
Finnish | 🇫🇮 |
| ✅ |
Hungarian | 🇭🇺 |
| ✅ |
Japanese | 🇯🇵 |
| ✅ |
Latvian | 🇱🇻 |
| ✅ |
Lithuanian | 🇱🇹 |
| ✅ |
Romanian | 🇷🇴 |
| ✅ |
Slovenian | 🇸🇮 |
| ✅ |
Swedish | 🇸🇪 |
| ✅ |
Thai | 🇹🇭 |
| ✅ |
Turkish | 🇹🇷 |
| ❌ |
Ukrainian | 🇺🇦 |
| ✅ |
Supported (may need fine-tuning)
Language | Code | Translate | |
Arabic | 🇪🇬 |
| ✅ |
Bengali | 🇧🇩 |
| ❌ |
Chinese | 🇨🇳 |
| ✅ |
Greek | 🇬🇷 |
| ✅ |
Hebrew | 🇮🇱 |
| ✅ |
Indonesian | 🇮🇩 |
| ❌ |
Maltese | 🇲🇹 |
| ✅ |
Marathi | 🇮🇳 |
| ❌ |
Norwegian | 🇳🇴 |
| ❌ |
Tagalog | 🇵🇭 |
| ❌ |
Tamil | 🇮🇳 |
| ❌ |
Telugu | 🇮🇳 |
| ❌ |
Vietnamese | 🇻🇳 |
| ❌ |
Language ID is weaker for Latvian and Slovenian and unavailable for Bengali, Marathi, Norwegian, Tamil, and Telugu. Audio Q&A was only evaluated in English.
Limitations
No word timestamps yet, so no SRT/VTT. The model can return per-word times and diarization segments, but this server doesn't request them yet.
Not yet exposed: vocabulary hints, nearest-speaker-only transcription, PII tagging.
Inputs are local files and YouTube links.
One job runs at a time.
Development
uv run pytest
uv run ruff check . && uv run ruff format --check .
npm install && npm run check:yamlTests need ffmpeg but not a model. CI runs the same checks on every pull request.
Available Tools
9 toolsask_about_audioA
Answer a question about one audio file or one YouTube link. The model was checked on English speech. A recording longer than the reasoning budget is refused. Pass start and end, in seconds, for the part you want. show_reasoning includes the model's reasoning before the answer. Waits wait_seconds (default 25). If it is still running, returns a job id. Tell the user, then call get_job_status. Do not poll in a loop.
Inputs: path (required), question (required), end, start, +2 more.
cancel_jobA
Stop a transcription job. Pieces that already finished stay readable through get_job_result.
Inputs: job_id (required).
get_job_resultB
The words from a finished job. offset and limit return a slice so a long transcript does not flood the chat. Only text is stored.
Inputs: job_id (required), limit, format, offset.
get_job_statusA
How far a transcription job has got. Pass wait_seconds from 0 to 45 to wait for it to finish. Tell the user the status. Do not poll in a loop.
Inputs: job_id (required), wait_seconds.
identify_languageA
Say which language is spoken in one audio file or one YouTube link. Returns one label, such as en_us. Listens to about 10 seconds of speech, skipping a pause at the start. Less reliable for Latvian and Slovenian. Not a transcript.
Inputs: path (required), label_set, candidates, clip_seconds, +1 more.
list_jobsB
Jobs on this machine, newest first. Use this to recover a job id.
Inputs: limit, state.
list_languagesA
Languages this model can handle, and how strong that support is. level is leading, production_ready, or supported. translate is yes when that language can be one side of an English translation. identify is yes, low, or no. Latvian and Slovenian are low. Norwegian is listed, but the card only scored transcription. It did not mark it supported.
Inputs: no named parameters.
transcribeA
Speech to text for one audio file or one YouTube link. style verbatim, the default, keeps every word, including fillers. style clean drops fillers and accidental repetitions and keeps the other words. style technical keeps emails, URLs, file paths, commands, and IDs exactly. Do not pass speakers with style technical. style raw_dictation drops punctuation and writes numbers as spoken words. A long file is cut at a pause and sent one piece at a time. Pass speakers from 2 to 8 to repeat 10 seconds at each join. Speaker numbers are matched across those joins. Waits wait_seconds (default 25). If it is still running, returns a job id. Tell the user, then call get_job_status. Do not poll in a loop.
Inputs: path (required), style, speakers, wait_seconds.
translateA
Write the speech in one audio file, or one YouTube link, in another language. source_language is the language spoken in the file, such as Polish or pl. target_language is the language the words should be written in, such as English or en. Pass both when you know them. A language passed here overrides NEMOTRON_SOURCE_LANGUAGE and NEMOTRON_TARGET_LANGUAGE. If source_language is empty, the language heard in the file overrides NEMOTRON_SOURCE_LANGUAGE. Leave target_language empty to use NEMOTRON_TARGET_LANGUAGE. One of the two languages must be English. A long file is cut at a pause and sent one piece at a time. Waits wait_seconds (default 25). If it is still running, returns a job id. Tell the user, then call get_job_status. Do not poll in a loop.
Inputs: path (required), wait_seconds, source_language, target_language.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 9 tools
Each tool has a clearly distinct action: transcribe, translate, ask_about_audio, identify_language, and list_languages operate on audio but produce different outputs, while get_job_result, get_job_status, list_jobs, and cancel_job cover distinct job-management roles. The descriptions make the boundaries explicit, so an agent should not confuse them.
All names use snake_case and start with a verb, with get_job_result, get_job_status, list_languages, list_jobs, cancel_job, ask_about_audio, and identify_language following a verb_noun pattern. transcribe and translate are bare verbs rather than verb_noun, a minor inconsistency.
Nine tools is a well-scoped set for transcription, translation, audio QA, language identification, and job management. Each tool has a clear role and none feels redundant.
The surface covers the core lifecycle: submit jobs (transcribe/translate), inspect status, retrieve results, list/cancel jobs, and query language support. Minor gaps include no job deletion or audio metadata tool, but these are not blocking.
Maintenance
Related MCP Connectors
Transcribes audio and video files to text, SRT, VTT and timestamped chunks for RAG.
Transcribe audio and video to text, SRT or VTT. Summaries, translation. Optional OAuth sign-in.
Transcribe audio and video files to text with timestamps and SRT or WebVTT subtitles
Whisper speech to text for podcast, audio and video URLs, 90+ languages, timestamped output.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables intelligent transcription of YouTube videos with automatic optimization for any video length, using local OpenAI Whisper processing and speaker diarization.-
- AlicenseNot gradedqualityCmaintenanceEnables transcription and speaker diarization of audio files, interviews, and YouTube URLs, producing speaker-attributed transcripts with timestamps. Supports multiple backends (local Whisper, OpenAI API) and output formats (txt, vtt, srt, json).Apache 2.0
- AlicenseNot gradedqualityCmaintenanceEnables local transcription of audio/video files and YouTube URLs, generation of SRT/VTT subtitles, and analysis of speech pacing and audience retention risk using faster-whisper.MIT

transcript1official
AlicenseNot gradedqualityBmaintenanceEnables AI clients to transcribe YouTube links and uploaded audio or video files, then retrieve the resulting transcripts as text, SRT, VTT, or JSON with timestamps and optional speaker labels. Also supports paging through long transcripts, checking upload progress, listing or deleting jobs, and viewing remaining account minutes.126 npm1MIT