What challenges did ASEAN Business Advisory Council face before AI implementation?

Provide seamless real-time multilingual understanding across 29 sessions for more than 2,300 delegates. Capture diplomatic nuance with very low latency and make post-event transcripts searchable and useful for research.

What was the impact of the AI solution?

Achieved sub-second translation latency instead of prior 10-20 second delays. Reached an 8.17/10 average satisfaction rating and reduced friction in diplomatic exchanges. Turned ephemeral speech into a searchable knowledge base.

Rank#1,083of 3,107

Source

GCP

ASEAN Business and Investment Summit (ABIS 2025): Real-time translation + Gemini Ultra post-event research

Use case typeMultilingual communicationUpdated Jan 1, 2025

ASEAN-BAC Malaysia and ABIS 2025 used Google Cloud to provide low-latency multilingual communication across 29 summit sessions for more than 2,300 delegates. The implementation combined Speech-to-Text, Cloud Translation API, Text-to-Speech, Compute Engine, Cloud Load Balancing, Firebase delivery, and Gemini Ultra plus NotebookLM for transcript research and summarization.

Organization: ASEAN Business Advisory Council
Industry: Public Sector
Location: Malaysia
Published: January 2025

Reported outcomes

8.2 /10

average satisfaction ratingCustomer experience

1 secondstranslation latency−50%translation delay

Strategic outcomes

Customer experience & trustRemoved friction from live diplomatic exchangesOther strategic outcomeCreated a searchable knowledge base from summit transcripts

Primary read

Use case focus

Showing 2 of 2

1Multilingual communication
2Knowledge management

Provide seamless real-time multilingual understanding across 29 sessions for more than 2,300 delegates.
Capture diplomatic nuance with very low latency and make post-event transcripts searchable and useful for research.

Deploy a low-latency streaming translation system with Speech-to-Text, Translation, and Text-to-Speech.
Use Compute Engine and Cloud Load Balancing for a scalable backend and Firebase for delivery.
Use NotebookLM and Gemini Ultra to query and summarize the transcript archive into structured research outputs.

Technologies

Google Cloud Speech-to-Text Cloud Translation API Text-to-Speech AI NotebookLM Gemini Ultra Google Compute Engine Cloud Load Balancing Firebase

Achieved sub-second translation latency instead of prior 10-20 second delays.
Reached an 8.17/10 average satisfaction rating and reduced friction in diplomatic exchanges.
Turned ephemeral speech into a searchable knowledge base.

Architecture

A low-latency streaming translation system ingested interpreter feeds with Google Cloud Speech-to-Text API, translated and synthesized audio with Cloud Translation API and Text-to-Speech API, scaled through Google Compute Engine and Cloud Load Balancing, and delivered subtitles to users via Firebase. NotebookLM and Gemini Ultra were then used to query and summarize the transcript archive for post-event research and reporting.

Implementation partners1

Cloud Space

Sources & evidence1

Groundedness: 5/5Type: Customer StoryPublished: Jan 1, 2025Publisher: Google CloudEvidence: PrimaryConfidence: High

Primary source

AI-generated summary. Verify important details with the linked sources before relying on this case.

Explore related AI use cases

Customer evidence

Provider evidence

GCP provider page GCP ranked cases

Market context

Public Sector AI use cases Public Sector industry insights Malaysia country insights