Integration · Productivity

Connect Claude & ChatGPT to MusicGPT

Ask your AI to write lyrics, turn them into a song or make a sound effect with MusicGPT. It can also work on audio you already have: split it into stems, remove background noise, detect key and BPM, transcribe it, convert the format or master it against a reference track. Jobs run in the background, and the AI checks back for the finished audio links.

Connect MusicGPT free

Free plan, no credit card. Takes about a minute.

Try asking

“Write lyrics for an upbeat song about a summer road trip.”
The AI writes song lyrics from your prompt.
“Turn those lyrics into an indie pop song.”
It starts a music job with the prompt, lyrics and style, then checks the job until the audio links are ready.
“Split this track into vocals, drums and bass.”
It starts a stem separation job on the audio link and returns a link to each stem when it is done.

What your AI can do

Music and lyrics

Generate songs from a prompt, write lyrics, and extend, remix or make a cover of an existing song.

Voice and sound

Turn text into speech, change the voice in a recording, generate sound effects and search the voice library.

Audio processing

Split stems, remove noise, detect key and BPM, transcribe, convert formats and master a track against a reference.

“What key and BPM is this song in?”
It detects the musical key, the BPM and any key changes in the audio file.
“Read this paragraph aloud in a calm female voice.”
It searches the voice library if needed and turns the text into speech with the voice or gender you choose.
“Remove the background noise from my podcast recording.”
It runs noise removal on the audio file and gives you a link to the cleaned audio.

17 tools for MusicGPT

These tools are switched on when you connect. You can switch any of them off, or require your approval before it runs.

change_voiceChanges dataConvert the voice in an audio file (`audio_url`) to another voice (`voice_id`), with optional pitch shift. Runs in the background and uses API credits.
convert_audio_fileChanges dataConvert an audio file (`audio_url`) to another `target_format`, with optional sample rate and bit depth. Runs in the background and uses API credits.
create_cover_songChanges dataMake a cover of a song (`audio_url`) sung by another voice (`voice_id`). Runs in the background and uses API credits.
detect_key_and_bpmChanges dataDetect the musical key, BPM and key changes of an audio file (`audio_url`). Runs in the background and uses API credits.
extend_songChanges dataExtend a song (`audio_url`) after a given time `extend_after` (seconds), with optional prompt and lyrics. Runs in the background and uses API credits.
generate_lyricsWrite song lyrics from a text `prompt`.
generate_musicChanges dataGenerate a song from a text `prompt`, with optional lyrics, music style, title, voice and instrumental options. Runs in the background and uses API credits; check the result with get_conversion.
generate_sound_effectChanges dataGenerate a sound effect from a text `prompt`, with optional length. Runs in the background and uses API credits.
get_conversionGet the status and result (audio links) of a job by `conversion_type` and either `task_id` or `conversion_id` from the job you started.
list_voicesList all available voices, page by page.
master_audioChanges dataMaster an audio file (`audio_url`) to match a reference track (`reference_audio_url`). Runs in the background and uses API credits.
remix_songChanges dataRemix a song (`audio_url`) following a text `prompt`, with optional new lyrics. Runs in the background and uses API credits.
remove_noiseChanges dataRemove background noise from an audio file (`audio_url`). Runs in the background and uses API credits.
search_voicesSearch the voice library by name with `query`. Use a voice ID for covers, voice changes or speech.
separate_stemsChanges dataSplit an audio file (`audio_url`) into stems such as vocals, drums and bass (`stems`). Runs in the background and uses API credits.
text_to_speechChanges dataTurn `text` into spoken audio, with an optional voice or gender. Runs in the background and uses API credits.
transcribe_audioChanges dataTranscribe an audio file (`audio_url`) to text, with optional language, translation and word timestamps. Runs in the background and uses API credits.

Set up in three steps

1
Pick the app
Create a free PipMCP account and choose MusicGPT from the app list.
2
Paste your key
Paste a MusicGPT API key. You need a MusicGPT API account: buy an API package at musicgpt.com/api. Then open the API Dashboard (musicgpt.com/api-dashboard) → API Keys → Create key, give it a name and click the eye icon to reveal and copy it.
3
Add the link to your AI
You get a personal MCP link. Add it to Claude, ChatGPT or Cursor:
1. Click your name, then Settings › Connectors › Add custom connector.
2. Paste your link as the Remote MCP server URL.
3. Switch it on from the + menu in a chat.

Questions

What can the AI do in MusicGPT?

It can generate music, lyrics, speech and sound effects, extend, remix and cover songs, and change the voice in a recording. It can separate stems, remove noise, detect key and BPM, transcribe, convert and master audio. It can also list and search voices and check the status and result of any job.

Does the AI see my MusicGPT credentials?

No. Your MusicGPT API key is encrypted at rest and never shown to the AI. After you save it, it is not shown again, not even to you. The AI only sees the results of the tools it calls.

Can I control what the AI is allowed to do?

Yes. You choose which tools are switched on, for example only lyrics and key detection. Generation and audio processing jobs use your MusicGPT API credits, so those tools can require your approval before they run. Every tool call is logged.

Does it work with ChatGPT?

Yes. In ChatGPT, open Settings › Apps & Connectors › Advanced, turn on Developer mode and add your PipMCP link. Developer mode needs a paid ChatGPT plan: Plus, Pro, Business or Enterprise. The same link also works in Claude (Settings › Connectors › Add custom connector), Cursor and other MCP clients.

What does it cost?

PipMCP has a free plan with no credit card required. Paid plans bill per completed task. You also need your own MusicGPT API account, and generation and audio jobs use your MusicGPT API credits.

How does the AI get my audio into MusicGPT?

The audio tools take a link to the file, so your recording needs to be reachable at a URL. The AI passes that link to MusicGPT, the job runs in the background, and the AI checks the job until the result is ready. You then get links to the finished audio.