Workshops Use Cases Architecture Projects Blog
About Contact Offers Talks
Home›AI technologies›Speech to textDeutsch

Speech to text: turning spoken words into text

Transcription, also called automatic speech recognition, turns audio into written text and records when each word was said.

Amazon TranscribeAzure AI SpeechSpeech-to-TextWhisper
Book a free 30-min call →AI Concept and Prototype workshops →

What is speech to text?

A speech recognition model listens to audio and writes down what was said. Good services also return a timestamp for every word, label who is speaking, and let you add your own vocabulary, so product names and abbreviations come out right.

The timings often matter more than the words. They are what lets you cut a video, write subtitles, or jump to a moment in a recording.

Goes inAn audio file, or a live stream
Comes outA transcript with word-level timings and speaker labels
Typical usesSubtitles, meeting notes, call analytics, searchable media archives

Four ways to run it

Each of the big clouds offers it as a managed service, and you can also run it yourself. Same task, a different bill and a different console.

Amazon Transcribe
Managed on AWS

Amazon Transcribe[1]

The managed speech-to-text service on AWS. You send audio and get back a transcript with word timings. AWS runs the model; custom vocabularies are yours.

Official documentation →
Azure AI Speech
Managed on Microsoft Azure

Azure AI Speech[2]

The managed speech service on Microsoft Azure, with speech to text in batch and in real time.

Official documentation →
Speech-to-Text
Managed on Google Cloud

Speech-to-Text[3]

The managed speech recognition API on Google Cloud.

Official documentation →
Whisper
Run it yourself

Whisper[4]

An open-weights speech recognition model from OpenAI, released under the MIT licence. It runs on your own machine or GPU, so the audio never has to leave.

Project page →
Amazon Transcribe: The managed service on one page, with the same service on the other platforms.
The managed service on one page, with the same service on the other platforms.[5]

How it looks in an architecture

Slides from my course on AI architectures at FH Burgenland, drawn from real projects.

One clip traced through the video pipeline: Transcribe writes the transcript that the subtitles and the agents work from.
One clip traced through the video pipeline: Transcribe writes the transcript that the subtitles and the agents work from.[5]
Five ready-made AI services, the technology underneath each one, and what it is called elsewhere.
Five ready-made AI services, the technology underneath each one, and what it is called elsewhere.[6]

Where I used it

The video pipeline

In my video pipeline, Transcribe writes the transcript behind the subtitles, for about 11 cents a clip. A custom vocabulary was the one setting that mattered: without it the product names came back wrong and every step after it inherited the mistake.

Read the write-up →

Managed and self-hosted together: hybrid

With a ready-made service the call leaves your network. The hybrid decision is what crosses, and in what state: the whole file, a redacted one, or only the part that has to. Where a file may not leave at all, the same step runs with the open-source option inside your building, and only the result, or nothing, crosses.[7]

The enterprise version of the video pipeline: the archive stays on the customer's NetApp ONTAP storage, and the AI services read it through S3 access points.
The enterprise version of the video pipeline: the archive stays on the customer's NetApp ONTAP storage, and the AI services read it through S3 access points.[5]

What would help you next?

Learn to decide

The course “From Managed to Hybrid” teaches where AI should run: what each layer costs, what you operate and where your data has to stay. It is in preparation, with a free live session first.

Join the course waitlist →
Linda MohamedLinda Mohamed
AI and cloud consultant, AWS Community Hero, Vienna

Build it with me

Tell me what you want to build. In 30 minutes we check which of the four ways fits your data, your rules and your budget, and whether AWS funding for a proof of concept can cover part of it. How AWS funding works →

Book a free 30-min call →15-min question slot

You already know what you need

Book the workshop directly:

ai-solutions.wiki

Read more on ai-solutions.wiki

My open knowledge base, with longer comparisons, pricing notes and sources.

Sources

  1. Amazon Transcribe documentation. https://docs.aws.amazon.com/transcribe/latest/dg/what-is.html
  2. Azure AI Speech documentation. https://learn.microsoft.com/en-us/azure/ai-services/speech-service/overview
  3. Speech-to-Text documentation. https://cloud.google.com/speech-to-text/docs
  4. Whisper. https://github.com/openai/whisper
  5. Linda Mohamed, course slides “The projects”, MASE-3, FH Burgenland, September 2026
  6. Linda Mohamed, course slides “How to run AI, layer by layer”, MASE-3, FH Burgenland, September 2026
  7. S3 access points on FSx for NetApp ONTAP, AWS documentation. https://docs.aws.amazon.com/fsx/latest/ONTAPGuide/s3-access-points.html
  8. Amazon Transcribe, ai-solutions.wiki. https://ai-solutions.wiki/tools/amazon-transcribe/
  9. Azure Speech Services, ai-solutions.wiki. https://ai-solutions.wiki/tools/azure-speech-services/
  10. OpenAI Whisper, ai-solutions.wiki. https://ai-solutions.wiki/tools/whisper/
  11. AWS AI services vs Azure AI, ai-solutions.wiki. https://ai-solutions.wiki/comparisons/aws-vs-azure-ai/

Links checked September 2026. Product names and features change: check the linked pages before you decide.