Inconsistent summarization quality with Read analyzeText (summarize=true) — how to improve accuracy/consistency?

[aygoel13]

Hi,

• We’re using Deepgram Text Intelligence (Read) to summarize call transcripts (agent + customer).
• We’ve observed inconsistent results: sometimes the summaries are accurate and useful, other times the same transcript produces a low-quality or inaccurate summary.
• We’d like to understand potential reasons for this variability and how to increase accuracy/consistency.

Sample Code snippet

const { result, error } = await deepgram.read.analyzeText(
{ text }, // full call transcript (agent + customer)
{
language: "en",
summarize: true,
}
);

Input -
Text: Full call transcript containing both agent and customer turns.
Language: en

Observed behavior -
Running summarize on the same transcript sometimes yields high-quality, on-point summaries.
Other times, the output misses key points or is noticeably less accurate, even with the same input text.

Expected behavior -
More consistent summarization quality across runs for identical inputs, or guidance on how to configure/use the API to improve consistency and accuracy.

Thank you !
Any guidance on stabilizing/improving summarization would be greatly appreciated.

[deepgram-community[bot]]

Thanks for asking your question. Please be sure to reply with as much detail as possible so the community can assist you efficiently.
_{Consider joining our Discord community for more opportunity to engage with your fellow Deepgram users. You can earn points which can be redeemed for cool stuff by being active in our communities!}

[deepgram-community[bot]]

Hey there! It looks like you haven't connected your GitHub account to your Deepgram account. You can do this at https://community.deepgram.com - being verified through this process will allow our team to help you in a much more streamlined fashion.

[deepgram-community[bot]]

It looks like we're missing some important information to help debug your issue. Would you mind providing us with the following details in a reply?

• The deepgram product you are using (e.g Speech to Text, Agent API)
• A request ID that triggered your error or issue.

[aygoel13]

We are using "Text Intelligence" for summarization.
Request id - 08930771-98ed-4c01-b6c0-8124e2a6b7cb

[aygoel13]

Hi,

Any update on this item ?

Thanks in advance.