All posts

こんにちは We’re Releasing an Enhanced Japanese (beta) Speech-to-Text Language Model

こんにちは We’re Releasing an Enhanced Japanese (beta) Speech-to-Text Language Model

We’re excited to now offer a speech-to-text language model for Japanese! Read on for details, including how to get access.

Did you know Japanese is spoken by over 125 million people worldwide? That's a lot of folks! Clearly, there's a lot of latent demand for a spoken-Japanese speech model. And that's why we're so excited to announce a beta rollout our Enhanced Japanese model which supports for even more EU languages.

Try Deepgram's Japanese Language Model

If you want to try out our Enhanced Japanese speech-to-text model you can quickly create an account on Deepgram Console and we’ll give you $150 in free credits. Simply select Japanese from the language menu when trying out our APIs.

If you’re already a Deepgram customer, you can call the Enhanced Japanese model using the following arguments:

  • model=general

  • version=beta

  • language=ja

  • tier=enhanced

You can see an example of a full API call using these arguments below.

Example API call

curl \
  --request POST \
  --header 'Authorization: Token YOUR_DEEPGRAM_API_KEY' \
  --header 'Content-Type: audio/wav' \
  --data-binary @youraudio.wav \
  --url 'https://api.deepgram.com/v1/listen?language=ja&tier=enhanced&version=beta&model=general'

What Can Developers Do with a Japanese Speech-to-Text Language Model?

  • Pair it with a Phone Call model to transcribe recordings from your European call center.

  • Pair it with a Meetings model to understand the topics commonly discussed by customers.

  • Create an Agent Assist solution to increase the productivity of your European sales team.

Key Benefits of Deepgram's Japanese Language Model

  • Accurately transcribe Japanese speakers with our Phone Call, Meeting, Voicemail, and Conversational AI use case models.

  • Many developers see upwards of 90% accuracy, depending on their use case.

  • Available for pre-recorded and streaming (getting started guide).

  • Transcribe on-premises or through the Deepgram Cloud.

  • Use in conjunction with speech understanding features such as Diarization (docs), Summarization, Topic Detection, and more.

There's More to Come!

Deepgram is the only speech AI provider to offer model customization and on-premises deployment options. We’ll continue to support the expansion of our language capabilities in the coming months to support our growing global customer base.

To learn more about the dozens of other languages and use cases Deepgram enables, please see the Language page in our documentation. And if you'd like to speak with someone about your project for diving in, please reach out! We welcome your feedback, please share it with us at https://www.deepgram.com/feedback.

If you have any feedback about this post, or anything else around Deepgram, we'd love to hear from you. Please let us know in our GitHub discussions .

More with these tags:

Share your feedback

Thank you! Can you tell us what you liked about it? (Optional)

Thank you. What could we have done better? (Optional)

We may also want to contact you with updates or questions related to your feedback and our product. If don't mind, you can optionally leave your email address along with your comments.

Thank you!

We appreciate your response.