adjust_volume
ChatGPTAdjust the volume of an audio track. Volume is a multiplier: 0.5 = half volume, 1.0 = unchanged, 2.0 = double volume (clamped to 0.01-5.0). Runs synchronously and returns the URL of the adjusted audio file. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
analyze_audio
ChatGPTAnalyze an audio file and return technical metadata: duration, sample rate, channels, bitrate, codec, and format. Use before mixing or editing to understand the source material. Runs synchronously. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
change_pitch
ChatGPTShift the pitch of audio without changing tempo. Semitones: positive = higher pitch, negative = lower pitch, range -12 to +12 (one octave). Runs synchronously and returns the URL of the pitch-shifted audio file. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
change_tempo
ChatGPTChange the speed/tempo of audio without affecting pitch. Two modes: (1) BPM mode — provide current_bpm and target_bpm, the rate is calculated automatically (use detect_bpm first if the current BPM is unknown). (2) Rate mode — provide rate as a multiplier (0.5 = half speed, 2.0 = double speed; clamped to 0.25-4.0). Runs synchronously and returns the URL of the retimed audio file. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
concatenate_audio
ChatGPTJoin 2-10 audio files end-to-end in sequence — combine an intro + verse + chorus, or stitch multiple clips together. Runs synchronously and returns the URL of the combined audio file. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
convert_format
ChatGPTConvert audio to a different format: aac (256kbps), flac (lossless), ogg (192kbps), or opus (128kbps). For wav/mp3/aiff output use run_file_converter instead. Runs synchronously and returns the URL of the converted file. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
convert_voice
ChatGPTConvert one or more vocal/audio tracks to a different AI voice from Vocuno's voice catalog. Requires at least one source audio URL and at least one voice_model_id from list_voices. Only use audio the user owns or has permission to use. SYNCHRONOUS and slow: each conversion takes roughly 1-2 minutes and the call waits for all of them; keep it to 1-2 targets per call. Each completed conversion returns a converted_url and a run_id (also retrievable later via get_run). The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Cost: 20 credits per audio minute for each conversion (each track × voice combination bills separately).
dereverb_audio
ChatGPTClean up audio: cleanup_type dereverb removes reverb, denoise removes background noise, deecho reduces echo/room reflections. Async: returns a task_id — poll get_dereverb with it every ~15 seconds; when complete, output_url is the cleaned audio. Source is either a direct audio_url or task_id + audio_id from a generated song. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Cost: 20 credits per call. Only use audio the user owns or has permission to use.
detect_bpm
ChatGPTDetect the tempo (BPM) of an audio track. Returns an integer BPM between 60 and 200. Use when the user asks for the tempo, or before change_tempo when the current BPM is unknown. Runs synchronously. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
fade_audio
ChatGPTAdd smooth fade-in and/or fade-out effects to an audio file. At least one of fade_in_seconds or fade_out_seconds must be > 0; musical defaults are about 2s for a fade-in and 6s for a fade-out. Runs synchronously and returns the URL of the faded audio file. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
generate_song
ChatGPTGenerate an original AI song with Vocuno. Simple mode: describe the song in prompt (genre, mood, topic, language). Custom mode: provide your own lyrics (and optionally style + title). Returns a task_id immediately; poll get_song until the song is complete (~1-3 min). Cost: 50 credits per song from the connected Vocuno account.
get_credits
ChatGPTGet the connected Vocuno account's remaining credit balance.
get_dereverb
ChatGPTCheck the status of an audio-cleanup task started with dereverb_audio. Poll every ~15 seconds until status is "complete" — output_url is then the cleaned audio and original_vocals_url the original upload (for A/B comparison). Free lookup.
get_run
ChatGPTThe status poller for every run_ tool (and for convert_voice results): fetch a run's status, step, progress, tracks (with audio_url/image_url), and any error. Poll every ~15 seconds until status is "complete" — or "error", in which case stop polling and explain the failure. Use ONLY a run_id returned by a run_ tool or convert_voice; song task_ids belong to get_song, stem task_ids to get_stem, dereverb task_ids to get_dereverb (the tool tells you which lookup to use if you mix them up). Free lookup.
get_song
ChatGPTCheck the status of a song generation started with generate_song. Tracks get a permanent share_url on vocuno.com — often already while still generating, as a streaming preview you should hand to the user immediately (the same link serves the final version once complete). Always share that link, never any raw file URL. Poll every ~15 seconds until status is "complete"; the cover art then comes attached as an image.
get_stem
ChatGPTCheck the status of a stem separation started with separate_stems and retrieve the per-stem audio URLs when complete. Poll every ~15 seconds until status is "complete"; the result then contains a stems map of stem name → audio URL, which can be fed into any other audio tool (mix_tracks, convert_voice, …). Free lookup.
highpass_filter
ChatGPTApply a highpass filter — cuts frequencies BELOW the cutoff. Use to remove low-end rumble (~80Hz), clean up muddy mixes (~100-200Hz), or create thin/telephone effects (~300Hz). Higher cutoff = more aggressive bass removal. Runs synchronously and returns the URL of the filtered audio file. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
list_library
ChatGPTList the connected account's recent Vocuno items with the audio_url values other tools accept: generated songs (task_id + tracks[].id, usable as task_id + audio_id), tool runs (extend, mashup, voice conversion, mastering, … with run_id) and stem separations (task_id + per-stem URLs). Newest first. Use it when the user refers to "my song", "the last track", "the stems from earlier", or gives a title instead of a URL. Free — costs no credits.
list_voices
ChatGPTList the AI singing voices available for convert_voice from Vocuno's built-in voice catalog. Returns up to 5 matching voices (id, name, gender, genres) plus the total match count; refine with the gender/genre/name filters to see other matches. The returned voice id is required for convert_voice. Free lookup.
loop_audio
ChatGPTLoop/repeat an audio clip. Specify either a number of repetitions or a target total duration in seconds. Runs synchronously and returns the URL of the looped audio file. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
lowpass_filter
ChatGPTApply a lowpass filter — cuts frequencies ABOVE the cutoff. Use to tame harshness/sibilance (~8-12kHz), create muffled/underwater effects (~2-4kHz), or pair with highpass_filter for a telephone effect (~3400Hz). Lower cutoff = more aggressive treble removal. Runs synchronously and returns the URL of the filtered audio file. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
mix_tracks
ChatGPTMix/overlay 2-5 audio tracks together at specified volumes — e.g. combine vocals with an instrumental, or layer separated stems back together. Runs synchronously and returns the URL of the new mixed audio file. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
render_music
ChatGPTDisplay playable Vocuno music cards when the user wants to browse or listen. Use source=library for recent songs, audio results, or stems; reuse the kind/query from list_library. Use source=song and an exact task_id from list_library or generate_song/get_song for a selected song and all its versions. Reads the connected account again; never pass invented IDs or media URLs. This does not generate, publish, or change music. Use after generation completes; keep get_song for status polling. Clients without UI receive a text summary.
run_audio_mastering
ChatGPTApply Vocuno mastering to an audio track for louder, clearer, wider, release-ready polish, saved as an Audio Mastering run. Runs synchronously and returns run_id; the mastered file URL is in get_run tracks[].audio_url. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Cost: 30 credits per call.
run_bpm_detector
ChatGPTDetect the tempo/BPM of an audio file and save the result to the user's Vocuno library as a BPM Detector run. Runs synchronously and returns run_id and bpm. Use detect_bpm instead when the user just wants the number without a library entry. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
run_extend_song
ChatGPTExtend an existing track with Vocuno: make it longer, add an intro/outro, or continue from a specific timestamp (continue_at seconds). Simplest valid call: audio_url + prompt (a missing prompt defaults to continuing in the same style). Pass lyrics + style only when specific words must be sung. Only use audio the user owns or has permission to use. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Returns run_id immediately — poll get_run with it every ~15 seconds until status is "complete" (or "error"). Cost: 75 credits per extension.
run_file_converter
ChatGPTConvert an audio file to wav, mp3, flac, aiff, aac, ogg, or opus and save the result to the user's Vocuno library. Runs synchronously and returns run_id + the output filename; the converted file URL is in get_run tracks[].audio_url. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
run_mixing
ChatGPTApply one server-side mixing/effect preset to an audio track and save it as a Mixing run: bass boost/reduction, treble boost, warm vintage, vocal clarity, underwater, cathedral reverb, slapback echo, telephone, FM radio, vinyl lo-fi, or loudness boost. Runs synchronously and returns run_id; the processed file URL is in get_run tracks[].audio_url. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Cost: 8 credits per call.
separate_stems
ChatGPTSeparate audio into individual stems (vocals, instrumental, drums, bass, etc.). Async: returns a task_id — poll get_stem with it every ~15 seconds until status is "complete". Source is either a direct audio_url, or task_id + audio_id from a generated song (get_song). The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Presets: vocals-instrumental (default, 2 stems), vocals-drums-bass (3 stems), full-band (discounted bundle: vocals, instrumental, drums, bass, guitar, piano), or custom with requested_stems from: vocals, instrumental, drums, bass, keys, strings, woodwinds, backing_vocals. Guitar/piano requests automatically become the full-band bundle. Common synonyms are auto-mapped (brass/horns/sax → woodwinds, synth/pads → keys, percussion → drums); unsupported stem names are rejected before any credits are charged. Cost: vocals-instrumental 20 credits per audio minute; multi-stem presets and full-band 40 credits per audio minute. Only use audio the user owns or has permission to use.
trim_audio
ChatGPTTrim/cut audio to a specific time range (extract a chorus, verse, intro, or any section). Runs synchronously and returns the URL of the trimmed audio file. The audio_url must be a direct audio file URL: use a URL from a previous Vocuno tool result (get_song track audio_url, get_run tracks[].audio_url, get_stem stem URLs, get_dereverb output_url, or the url returned by an audio-editing tool), or a user-provided direct audio file URL. Never pass a YouTube/streaming/page URL — it will be rejected. For a LOCAL file call upload_audio first; to reuse an existing Vocuno song/run/stem call list_library. Free — costs no credits.
upload_audio
ChatGPTBring a LOCAL audio file into Vocuno so other tools can use it. This tool does NOT take the file contents: it returns a one-time upload_url (HTTP PUT, valid 30 minutes), the exact headers to send, a ready-to-run curl command, and the permanent audio_url the file will have. Flow: call upload_audio with the filename → run the returned curl command (or PUT the raw bytes with the returned headers) → pass the returned audio_url to separate_stems, trim_audio, convert_voice, run_extend_song, etc. Accepts audio (mp3, wav, flac, m4a, aac, ogg, aiff) and video (mp4, mov). Uploads are kept for 30 days. Free — costs no credits.
