Voice · Vendor site 2026-10-04
AudioCraft is Meta AI’s open research codebase for generative audio, unifying MusicGen, AudioGen, and EnCodec models trained on raw waveforms. The demo site explains tokenized audio modeling and links to code, model cards, and sample generations rather than a hosted consumer product.
Researchers and developers experimenting with text-to-music, text-to-sound, and neural audio codec models from Meta AI.
Capabilities require running the published codebase locally or in your own infrastructure; Meta does not sell a managed AudioCraft subscription on this site.
AudioCraft consolidates Meta AI audio models and documentation for researchers building music, SFX, and compression pipelines from raw audio signals.
What it can do
Unknown is different from unavailable. Each fact carries its own evidence.
| Capability | Value | Evidence | Checked |
|---|---|---|---|
| Unified codebase | Homepage describes AudioCraft as a single-stop codebase for generative music, sound effects, and compression models. | Facts sourced | 2026-10-04 |
| MusicGen and AudioGen | Site sections cover text-to-music via MusicGen and text-to-sound environmental generation via AudioGen. | Facts sourced | 2026-10-04 |
| EnCodec tokens | Overview explains EnCodec maps waveforms to discrete tokens consumed by autoregressive language models. | Facts sourced | 2026-10-04 |
Understand the total cost
Free
Free
Research site links to open code and model resources; no paid SaaS tier is offered on the AudioCraft homepage.
TTS.aiTTS.ai is a multi-model text-to-speech platform with 36+ models, 314+ voices, voice cloning, speech-to-text, music, and marketplace listings. The homepage offers free generation without an account while paid plans add commercial licenses, API access, and longer downloads.
Explore tool
MusicfyMusicfy is an AI music and voice platform for generating songs, covers, and voice models from user audio. Subscription tiers raise monthly song counts, upload limits, voice targeting, and commercial licensing.
Explore tool
DiffRhythmDiffRhythm.ai is a browser AI music generator built on latent diffusion research that creates full songs with vocals and accompaniment from lyrics and style prompts. The site advertises fast non-autoregressive generation up to several minutes, multiple model versions, and credit-based subscriptions plus a limited free quota.
Explore toolThe practical questions
AudioCraft is Meta AI’s open research codebase for generative audio, unifying MusicGen, AudioGen, and EnCodec models trained on raw waveforms. The demo site explains tokenized audio modeling and links to code, model cards, and sample generations rather than a hosted consumer product.
Research site links to open code and model resources; no paid SaaS tier is offered on the AudioCraft homepage.. This record lists ongoing free access; check the plan limits before starting.
No paid monthly price is listed; this record treats the product as free to start. See the plan cards for entitlements, billing commitments and seat minimums.
Not confirmed. API access and subscription access may have different terms; consult the linked sources.
Capabilities require running the published codebase locally or in your own infrastructure; Meta does not sell a managed AudioCraft subscription on this site.
No retained pricing changes yet. A current price alone does not establish a historical trend.
Reviewed vendor source
Read original source ↗Facts apply to the named version and check date. Send a sourced correction if something changed.