#Accent change for Voice Agent API

1 messages · Page 1 of 1 (latest)

cursive chasmBOT
#

Hi @dawn yoke. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:

light basaltBOT
#

Hey there! This channel allows you to ask Kapa.ai for assistance. But it looks like you haven't connected your Discord profile to your Deepgram account yet. You can do so at https://community.deepgram.com - by doing so, you'll be able to ask questions in #1115960287183511643 if you need human assistance!

Deepgram Community

Deepgram’s community offers tons of resources to support you on your voice AI journey

cursive chasmBOT
#

Based on the knowledge sources provided, there are a few approaches you can consider to address accent-related transcription issues with the Voice Agent API:

  1. Choose the right model: Deepgram offers various models optimized for different use cases. For general purposes, 'nova-2-general' is recommended, while 'nova-2-phonecall' is better for lower-quality or phonecall audio. There are also domain-specific models for specific fields that might better handle certain accents GitHub Discussion.

  2. Use Keyword Boosting: This feature allows you to specify words that Deepgram should pay special attention to during transcription. It's particularly useful for uncommon words, proper nouns, or terms that might be affected by accent variations GitHub Discussion.

  3. Consider Custom Model Training: If you're experiencing significant issues with accent recognition, Deepgram offers a "white glove" service to fine-tune models to your specific data. This is part of their Model Improvement Partnership Program, which can help increase accuracy for your specific domain and use case The Deepgram Model Improvement Partnership Program.

  4. Language Selection: Make sure you're specifying the correct language parameter in your API calls. The API defaults to English if no language is specified, which could cause issues with accented speech in other languages GitHub Discussion.

#

For the Voice Agent's text-to-speech output, you can select from various accented voices. Deepgram offers voices with different accents including US English, UK English, and Irish English Voice Selection.

If you're dealing with multilingual content (like Spanish with English words mixed in), you might want to use their Latin American Spanish model with the appropriate parameters GitHub Discussion.

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:

#

@dawn yoke Kapa.ai is still learning and improving, please let me know how I did by reacting below.