CassetteAI

Freemium

面向音乐、音效和语音的实时生成式音频平台

什么是 CassetteAI？

CassetteAI is a real-time generative audio product for music, sound effects, and speech. It presents those modalities through a single API and SDK, with an emphasis on low-latency output on edge hardware or via a hosted API for developers who do not have on-device access.

The site positions the product for production workflows where audio needs to be created inside an app rather than in a separate editing tool. Music and SFX are live, while text-to-speech is listed as coming soon. Pricing is metered per output minute or per generation rather than sold as seat-based plans.

CassetteAI 能做什么？

Prompted music generation

Generate adaptive music from prompts for moods, genres, tempos, keys, or references. The site says music tracks stream into the app while the user plays and can run for 1 to 4 minutes, with 44.1 kHz stereo output.

Sound effects on demand

Create sound effects from natural-language event descriptions such as door slams, power-ups, or ambience. CassetteAI says SFX can be loop-safe, per-frame re-rolled, and rendered in roughly 1 second for up to 30 seconds of audio.

Single SDK and API pattern

Use one API shape across modalities and swap the model ID between music, SFX, and TTS. The site shows `fal.subscribe()` examples in JavaScript, Python, and cURL.

Low-latency streaming

Support real-time output with low first-sample latency and streaming responses. The homepage cites a 23 ms first-sample latency and under-50 ms streaming responses on edge hardware.

Production audio format

Generate reference-grade audio output at 44.1 kHz stereo and download it as `.wav`. The site says this matches DAW expectations and keeps output consistent for production use.

使用场景

“Game and interactive app music”

Add adaptive background music to games or interactive apps, where the track needs to change with the session and stream into the experience while the user plays.

“Sound design for product events”

Generate short, specific sounds for UI actions, gameplay events, or media tools, including loop-safe ambient effects and one-off event sounds.

“Developer integration”

Use the API from application code in JavaScript, Python, or cURL to wire audio generation into an existing pipeline without moving to a separate studio workflow.

“Low-latency audio workflows”

Build real-time audio features that need low first-sample latency and fast turnarounds, such as live creator tools, accessibility tooling, or browser-based experiences.

“Planned speech generation”

Prepare for speech features by following the TTS waitlist and keeping the same API shape in mind for a future text-to-speech release.

常见问题

How do I use CassetteAI in a project?

CassetteAI exposes a single API for music, sound effects, and TTS. The site says the developer API works with JavaScript, Python, and cURL, and that the hosted API is available for developers without on-device access.

How fast is generation?

Music generation is described as returning a 30-second sample in under 2 seconds and a full 3-minute track in under 10 seconds, while SFX generation renders up to 30 seconds in roughly 1 second of processing time.

What does CassetteAI cost?

The site describes per-use billing: music at $0.02 per output minute and sound effects at $0.01 per generation. The pricing page also says there are no monthly commits or developer seats.

Can CassetteAI run on device?

The homepage and pricing text both indicate the product is designed to run on edge hardware or on device, with a hosted API also available. The about page says the models fit in your app bundle and emphasizes low-latency, streaming responses.

Is text-to-speech available now?

The site says the TTS model is ‘soon’ and references a waitlist, so music and SFX are the live modalities while text-to-speech is still launching.