bolna
bolna copied to clipboard
No Audio in Call
Hello, I am not able to here AI voice while running the "/call" API. But I am able to see the AI response on terminal logs. Can you please let me know what can be the issue here? Thanks.
have you setup ngrok auth token?
yes, I have setup ngrok auth token in "ngrok-config.yml" file.
@wdcs-divyeshpatel what synthesiser are you using?
I tried OpenAI, Deepgram and Polly. None of them worked.
Can you confirm if the audio is being synthesised?
Can you please give me steps on how to confirm it?
@wdcs-divyeshpatel We pushed few changes over the last 3-4 days. Can you please take the latest pull and try again.
Same issue here. I pulled the latest version
+1
hey @navjotdhanawat can you paste the agent config you used?
@prateeksachan Using same payload mentioned in readme for local setup.
{
"agent_config": {
"agent_name": "Alfred",
"agent_type": "other",
"agent_welcome_message": "Welcome",
"tasks": [
{
"task_type": "conversation",
"toolchain": {
"execution": "parallel",
"pipelines": [
[
"transcriber",
"llm",
"synthesizer"
]
]
},
"tools_config": {
"input": {
"format": "pcm",
"provider": "twilio"
},
"llm_agent": {
"agent_flow_type": "streaming",
"provider": "openai",
"request_json": true,
"model": "gpt-3.5-turbo-16k",
"use_fallback": true
},
"output": {
"format": "pcm",
"provider": "twilio"
},
"synthesizer": {
"audio_format": "wav",
"provider": "elevenlabs",
"stream": true,
"provider_config": {
"voice": "Meera - high quality, emotive",
"model": "eleven_multilingual_v2",
"voice_id": "TTa58Hl9lmhnQEvhp1WM"
},
"buffer_size": 100.0
},
"transcriber": {
"encoding": "linear16",
"language": "en",
"model": "deepgram",
"stream": true
}
},
"task_config": {
"hangup_after_silence": 30.0
}
}
]
},
"agent_prompts": {
"task_1": {
"system_prompt": "Ask if they are coming for party tonight"
}
}
}
@navjotdhanawat i just checked - it's working fine for me. Can you cross-check if the elevenlabs voice id works for your account? Or maybe you can put in a different voice id that's available for your account.
@prateeksachan the voice_id
i was using for elevenlabs was incorrect, created new voice and its working now. Thanks