Deepgram offers advanced transcription services including features like speaker diarization. This feature identifies different speakers within an audio stream and is particularly useful for multi-speaker environments like meetings or interviews.
Diarization is the process of partitioning an input audio stream into segments according to the identity of the speaker, ensuring that each speaker’s segments are unique and separate. This enhances the transcription by attributing the spoken content to the correct individual, making the output more structured and informative.
Diarization does not incur an additional cost beyond the regular transcription fees you pay with Deepgram. Transcription services, including diarization, are priced based on usage, measured in terms of audio requests and the duration of audio processed. While there are no additional charges for enabling diarization, all transcription features are effectively part of your overall usage, which determines your monthly billing.
This usage-based pricing applies to all transcription features, including:
With this model, users can take full advantage of these powerful features without worrying about separate costs for each function.
To best utilize these features within your budget, consider the specific needs of your use case. If your application benefits from distinguishing between multiple speakers, diarization will be an invaluable feature. For projects focusing on simple speech-to-text conversion without the need to separate speakers, you could opt-out of diarization to streamline your processing.
For more details on Deepgram’s pricing, visit the Deepgram Pricing Page. Should you have further questions or need assistance, feel free to reach out to the community through Deepgram Discord or post queries on Deepgram GitHub Discussions.
This article ensures that you are fully aware of the capabilities and cost-effectiveness of using Deepgram's full suite of transcription features, including diarization, without incurring hidden fees.