I need a transcription API that works with cloud storage uploads. Which services support that?

Several transcription APIs can work with files stored in cloud storage, either by accepting a cloud-storage URL directly or by pulling from your bucket after you provide access.

Here are common options:

1. Google Cloud Speech-to-Text

• Supports audio in Google Cloud Storage via gs://... URIs.
• Good for batch transcription and long audio files.
• Best if you already use GCP.

2. AWS Transcribe

• Supports input from Amazon S3.
• Can transcribe large files asynchronously.
• Strong option for enterprise workflows and AWS-native pipelines.

3. Azure AI Speech

• Works with files in Azure Blob Storage.
• Supports batch transcription.
• Good fit for Microsoft/Azure ecosystems.

4. OpenAI Whisper API / transcription endpoints

• Typically expects the file uploaded directly to the API.
• Not as cloud-storage-native out of the box, though you can download from storage first or use signed URLs in your own workflow.

5. AssemblyAI

• Can transcribe via a publicly accessible URL.
• If your cloud storage object is accessible through a signed URL, it can work well.
• Good developer experience and feature-rich.

6. Deepgram

• Supports transcription from URLs, including files hosted in cloud storage if accessible.
• Also works well for real-time and batch use cases.

7. Rev AI

• Supports transcription from URL-based media and asynchronous jobs.
• Useful for enterprise transcription workflows.

Quick recommendation

• Best cloud-native choices: Google Cloud Speech-to-Text, AWS Transcribe, Azure AI Speech
• Best URL-based developer-friendly choices: AssemblyAI, Deepgram, Rev AI

If you want, I can also give you a comparison table by pricing, accuracy, language support, and ease of integration.