Case Studies | SaaS B2B

Checking every take against the script

Audio File Transcription and Analysis

About

A SaaS B2B business working with Cloud Combinator on AWS. The client is anonymised at their request.

Challenge

The proof of concept set out to solve four things.

Matching audio to script

The core need was to transcribe each audio file and compare it to the original script. The first focus was an accurate similarity score between what was written and what was actually said.

Working across languages

Content is not only in English. The second focus was to translate non-English phrases with a large language model before comparison, so the same quality check works in any supported language.

Pinpointing every phrase

Knowing a phrase was spoken is not enough. The third focus was to capture the timestamps where each phrase occurs, including repeats, with a similarity score for each occurrence, output as a CSV.

Reliable at volume

Speech and language services have usage limits. The final focus was quota-aware orchestration so large batches process dependably without exceeding service limits.

Solution

Cloud Combinator delivered the solution as a set of AWS Step Functions workflows, deployed as infrastructure as code, that move each file safely from upload to a finished, queryable result.

Similarity

Per-phrase match between script and recorded audio

20+

Supported via Amazon Transcribe and Amazon Bedrock

Step

Quota-aware orchestration for reliable processing at volume

By the numbers:

  • Similarity - Per-phrase match between script and recorded audio
  • 20+ - Supported via Amazon Transcribe and Amazon Bedrock
  • Step - Quota-aware orchestration for reliable processing at volume
Changes

Cloud Combinator delivered a working proof of concept, approaching a production-grade service, that transcribes audio, translates where needed and scores it against the script phrase

  • Automated QA — Amazon Transcribe and Amazon Bedrock replace manual checking, scoring how closely each recording matches its script.
  • Language-agnostic — Non-English phrases are translated before comparison, so the same check works across the many languages the client handles.
  • Phrase-level detail — Every phrase is located with timestamps and a similarity score, including repeats, delivered as a ready-to-use CSV.
  • Built to scale reliably — Step Functions with quota-aware SQS queues keep large batches processing dependably within service limits, on a resilient, replicated storage layer.
  • Handover — Included deployment via CloudFormation in a secure VPC, documentation and a knowledge-transfer session.

By phrase. It was deployed via CloudFormation in the client's AWS environment and handed over with documentation and a session.

With the concept proven and the design already close to production, the client has a clear route to hardening the service, adding Bedrock guardrails, error handling and retries, and rolling it out across more regions and languages as demand grows.

AWS Stack

Amazon Transcribe

For converting audio recordings into text.

AWS Step Functions

For quota-aware orchestration of the transcription and comparison workflows.

AWS Lambda

And Amazon SQS for processing steps and throttling within service limits.

Amazon S3

For structured, replicated and lifecycle-managed storage of audio and results.

Amazon DynamoDB

For metadata, processing status and results, with Amazon API Gateway for upload and retrieval.

YOU MIGHT LIKE

Related success stories

View all case studies

Case Studies | Insights

Utilising Language Recognition, Speed, and Enhanced Security to Make Social Media a Force for Good

  • Here, we take a detailed look at how the Cloud Combinator team collaborated with another cutting-edge AI service provider that provides intelligent systems to “make social media more social” for brands and users alike.
  • Arwen AI is a UK-based startup specialising in AI solutions to manage and enhance brands’ social media interactions. Founded in 2020 by Matt McGrory, Dr. David Cole, and Joel Bailey, Arwen. AI focuses on using AI to automatically detect and remove spam, toxic comments, and other unwanted content from social media platforms.
  • The team at Arwen have three core products. ‘Moderate’ is focused on identifying and removing toxic content from social media channels. ‘Engage’ helps brands identify and engage with meaningful conversations on social media, and ‘Customize’ allows brands to apply bespoke algorithms to their channels - creating an even more effective moderation and engagement.
Read more
CONTACT US

Ready to turn AI into impact?

We'll help you spot the highest-value opportunities, reduce risk around your first AI initiative, and define a clear path to results from day one.

Why talk to us:

Outcome-driven recommendations

AWS-recognised delivery expertise

Risk-aware AI adoption

Clear next step, not a sales pitch

Start with a focused 20-minute conversation about your goals — no pressure, no commitment.

We use cookies

This website uses cookies to enhance user experience and to analyze performance and traffic on our website.

See our Privacy Policy for details.