Audio Encoding. Built for Developers.
Encode, merge, loop, extract artwork, and generate waveforms — all automated in a Simple API.
Create smaller audio files with control over quality and bitrate.
Automate audio processing with one reusable workflow.
Create transcripts, waveforms, and playback formats from one upload.
Build momentum with audio
Reach more listeners. Publish audio faster.
Give your engineers time to build. We handle the audio infrastructure.
Reach more listeners
Deliver audio in MP3, AAC, FLAC, ALAC, Opus and more for the devices your listeners use.
Publish audio sooner
Automate encoding, waveforms, transcripts, and exports in one upload.
Make audio easier to discover
Turn speech into searchable text. Translate transcripts to reach more listeners.
Welcome more contributors
Let users resume audio uploads even on slow, unreliable connections.
Handle traffic spikes of all sizes
Buy extra priority slots in advance. We queue jobs when your slots are busy.
Volume pricing for audio at scale
Transloadit is cheaper than AWS MediaConvert, Filestack, Cloudinary and Uploadcare at scale.
Transform Your Audio
At Transloadit, a Robot is an automated worker that performs a single task in your audio workflow — like extracting cover artwork, encoding audio, or concatenating audio files.
- /audio/encode robot
Encode Audio Files
Convert audio files into all kinds of formats. Like MP3, AAC, and more.
- /audio/artwork robot
Extract or Add Cover Artwork
Extract embedded cover artwork or add it to your audio files.
- /audio/concat robot
Concatenate Audio Files
Concatenate several audio files together into a single audio file.
- /audio/loop robot
Loop Audio Files
Loop one audio file as often as is required to match a given duration.
- /audio/merge robot
Merge Audio Files
Overlay several audio files on top of each other.
- /audio/split robot
Split Audio Files
Turn long recordings into clips or chapters with precise time ranges.
- /audio/waveform robot
Generate Waveforms
Generate waveform images for your audio files. Fully customizable styling.
- /speech/transcribe robot
Transcribe speech in audio files
Turn spoken words into searchable text.
- /text/speak robot
Convert text into speech
Write text in a document, and have it spoken aloud.
Build Audio Pipelines by Chaining Robots
Chain Robots together to create end-to-end automated pipelines for audio processing. Our AI assistant will write your Template for you.
{"steps": {":original": {"robot": "/upload/handle"},"encode_mp3": {"use": ":original","robot": "/audio/encode","preset": "mp3","ffmpeg": {"ab": "320k","ar": "44100"}},"artwork": {"use": "encode_mp3","robot": "/audio/artwork"},"waveform": {"use": "encode_mp3","robot": "/audio/waveform","style": "v1","width": 640,"height": 160,"waveform_style": "bars","bar_style": "rounded","bar_width": 6,"bar_gap": 4,"pixels_per_second": 54,"amplitude_scale": 0.7,"center_color": "f97316ff","border_color": "00000000"}}}
Your workflow. Your controls.
Audio workflows without the pain
Cut the cost and complexity of in-house audio processing. One API, your workflow.
You retain full control
Customize FFmpeg parameters directly. Go beyond presets with full control over codecs, bitrates, sample rates, channels, filters, and more.
Keep listeners immersed
Normalize loudness so listeners spend less time adjusting the volume between tracks.
Webhooks and real-time updates
Get per-file callbacks with /http/request (alpha), or an Assembly Notification when processing ends.
Secure uploads and delivery
Control upload access with expiring, server-generated signatures. Protect audio delivery with signed URLs.
Block malicious audio files
Add virus scanning to reject malware-infected audio files before further processing.
Control processing costs
Choose included usage and pay for overages. Set a soft bill limit for alerts or a hard limit to stop processing.
Turn speech into transcripts
Transcribe podcasts, interviews, and voice recordings in the same workflow that prepares your audio for playback.
Bring your own storage
Send finished audio directly to Amazon S3, Cloudflare R2, Google Cloud Storage, Azure, and more.
Multiple outputs. One API call.
Encode, transcribe, generate waveforms, and export all in the same one API call.
Prepare your audio for publishing
Frequently asked questions
How do I encode my first audio file?
Choose an audio preset, save a Template, and try a sample upload in the Template editor. Once the result sounds right, reuse that Template through Uppy or an SDK. Explore audio encoding options.
Which audio formats and settings can I use?
Working with MP3, WAV, AAC, FLAC, Ogg, or Opus? We support these popular audio formats and many more. Start with a preset or customize FFmpeg parameters for codecs, bitrate, sample rate, channels, and filters. Browse supported audio formats.
Can I normalize loudness and extract audio from video?
Yes. The audio encode Robot accepts audio and video inputs. Choose an audio preset to extract a video’s audio track, or use FFmpeg filters to normalize loudness. Test your settings on representative recordings. See audio encoding controls.
Can I create waveforms and transcripts in the same workflow?
Yes. Connect audio encoding, waveform generation, and speech transcription steps in one Assembly. Export the results together so your app can show a player, waveform, and transcript. Explore waveform generation.
Can listeners read or search a transcript?
Use the speech transcription Robot to create text from podcasts, interviews, and recordings. Your app can display or index that text, or pass it to a separate text translation step to reach more languages. Language and output options depend on the provider, and transcripts may need review. See speech transcription options.
Can users upload large recordings on unreliable connections?
Yes. Resumable uploads let users continue after a connection drops instead of restarting the whole recording. Maximum file size depends on your plan; contact us about larger requirements. Check current upload limits.
Can I keep audio in my own storage and get completion updates?
Yes. Export audio to Amazon S3, Cloudflare R2, Google Cloud Storage, Azure, and more. Use Assembly Notifications when processing finishes, or the HTTP request Robot (currently in alpha) for callbacks during processing. Set payload to metadata to identify each file. Your endpoint must return 2xx before the request timeout, or the step fails. Explore HTTP callbacks or learn how to save your results.
How can I estimate and control costs?
Rates vary by operation and usage. Set a soft bill limit for alerts or a hard limit to stop processing. For larger workloads, ask about volume pricing and extra priority slots. Estimate your workflow costs.