Back office ops · Production

Delphi leverages AssemblyAI to expand language support 16x and cut clone training time by 50%

The problem

Delphi's clone training pipeline needed to process diverse audio and video content in many languages, creating a bottleneck in accurately transcribing multilingual material to train AI clones at scale.

Workflow diagram · grounded in source
1
Diverse content ingestion
trigger
“The content used to train these digital clones often comes in various formats, including audio and video”
2
Automatic language detection
ai_action
“Automatic Language Detection (ALD)”
3
Speaker diarization and timestamping
ai_action
“High-quality Speaker Diarization (speaker labeling) with accurate timestamps”
4
Multilingual transcription
ai_action
“accurately detect and transcribe content in a much wider range of languages”
5
Clone training pipeline
ai_action
“This integration enables Delphi to rapidly analyze and process diverse content types, extracting vital details about the clone creators' material. The result is a significant boost in the speed and accuracy of Delphi's digital twin creat…”
6
Credibility through citations
output
“This precision in content referencing, enabled by accurate speaker labeling and timestamps, adds a new dimension of credibility and depth to interactions with Delphi's digital clones”
Reported outcome

After integrating AssemblyAI, Delphi expanded language support by 16x, reduced clone training time by 50%, and added a precise citation feature enabled by accurate speaker labeling and timestamps.

Reported metrics
Language support expansion16x
Clone training time50%
Reported stack
AssemblyAIAutomatic Language DetectionSpeaker Diarization
Source
https://www.assemblyai.com/customers/delphi-customer-story
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

After integrating AssemblyAI, Delphi expanded language support by 16x, reduced clone training time by 50%, and added a precise citation feature enabled by accurate speaker labeling and timestamps.

What tools did this team use?

AssemblyAI, Automatic Language Detection, Speaker Diarization.

What results were reported?

Language support expansion: 16x; Clone training time: 50% (source-reported, not independently verified).

How is this back office ops AI workflow structured?

Diverse content ingestion → Automatic language detection → Speaker diarization and timestamping → Multilingual transcription → Clone training pipeline → Credibility through citations.