https://retellai.com/?utm_source=discord&utm_medium=seo logo
how can I make it so that the prompt/agent listens...
# ❓・ai-evy-support
p
@AI Evy how can I make it so that the prompt/agent listens better? I believe my agent is having issues hearing me
a
Thanks for tagging me @papakwame_12. I'll look through relevant sources and get back to you shortly.
### Answer If your agent is having trouble hearing you, there are several configuration options you can adjust to improve speech recognition: 1. Add Boosted Keywords Configure
boosted_keywords
to bias the transcriber model toward specific words or terms that are commonly used in your conversations [(1) ]() [(2) ]() [(3) ]() . This is especially helpful for names, brands, or domain-specific terminology:
Copy code
json
"boosted_keywords": [
  "retell",
  "kroger"
]
[(1) ]() 2. Adjust STT Mode for Accuracy Set
stt_mode
to
"accurate"
instead of
"fast"
to prioritize transcription quality over speed :
Copy code
json
"stt_mode": "accurate"
3. Configure Denoising Mode Use
denoising_mode
to filter background noise [(1) ]() [(3) ]() . The default is
"noise-cancellation"
, but you can also try
"noise-and-background-speech-cancellation"
for environments with multiple speakers:
Copy code
json
"denoising_mode": "noise-cancellation"
[(1) ]() 4. Set the Correct Language Ensure the
language
parameter matches the actual spoken language [(1) ]() [(4) ]() . Mismatched settings significantly increase transcription errors [(4) ]() :
Copy code
json
"language": "en-US"
[(1) ]() 5. Add Custom Pronunciation Use
pronunciation_dictionary
to guide audio synthesis for consistent pronunciation of specific words [(1) ]() [(3) ]() :
Copy code
json
"pronunciation_dictionary": [
  {
    "word": "actually",
    "alphabet": "ipa",
    "phoneme": "ˈæktʃuəli"
  }
]
[(1) ]() 6. Adjust Custom STT Configuration If you need more control, configure
custom_stt_config
with specific provider settings like endpointing timing :
Copy code
json
"custom_stt_config": {
  "provider": "azure",
  "endpointing_ms": 123
}
High-quality ASR (Automatic Speech Recognition) is critical because if transcription is inaccurate, even advanced AI systems will misunderstand user intent [(5) ]() . Strong ASR delivers faster, more accurate conversations and better intent recognition .
​ To learn more, see the sources I considered: [[1] Create Voice Agent](https://docs.retellai.com/api-references/create-agent) [[2] Speech Recognition Quality Issue](https://discord.com/channels/1186424376300679209/1224173869758414899) [[3] Discord message from @immaculate_cheetah_77268 on 2025-09-04](https://discord.com/channels/1186424376300679209/1186462578310910112/1413112256220368896) [[4] Issues with Portuguese (PT-PT) Voice Agent: Latency, Audio Recognition, and Email Capture](https://community.retellai.com/t/issues-with-portuguese-pt-pt-voice-agent-latency-audio-recognition-and-email-capture/711) [[5] Automatic Speech Recognition (ASR)](https://www.retellai.com/glossary/automatic-speech-recognition-asr) ​
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.
p
@AI Evy can i find all you recommended on the dashboard?
a
Thanks for the follow-up, @papakwame_12. I'll get back to you soon.
### Answer Yes, most of the settings I recommended can be found in the dashboard under your agent's configuration settings. In the Dashboard, you'll find: Speech Settings section: - **Boosted Keywords**: Provides biases toward certain words, making them easier to recognize - commonly used for brand names, people's names, etc. [(1) ]() [(2) ]() - **Speech Normalization**: Converts entities like date, currency, numbers into plain words - **Pronunciation**: Set up pronunciation guides for specific words - **Responsiveness**: Controls how responsive the agent is - **Interruption Sensitivity**: Controls how fast the agent gets interrupted by user interruptions - **Backchanneling**: Set up how often and what words the agent uses to acknowledge users - **Reminder frequency**: How often the agent will remind the user when user is inactive Language Settings: - **Select Language of Agent**: Select what language the agent will understand (the language of user audio) [(2) ]() Call Settings section: - **Voicemail related settings**: Set up voicemail detection and what to do when voicemail is detected [(1) ]() [(2) ]() According to community discussions, you can find these settings by accessing your agent settings from the dashboard [(3) ]() . The STT mode and denoising mode settings may require API configuration or may be available in more advanced settings sections of the dashboard.
​ To learn more, see the sources I considered: [[1] Step 2: Configure the basic settings](https://docs.retellai.com/build/single-multi-prompt/configure-basic-settings) [[2] Step 1: Configure global settings](https://docs.retellai.com/build/conversation-flow/global-setting) [[3] Change the STT](https://discord.com/channels/1186424376300679209/1466175091724255372) ​
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.