Services Overview

Daily Bots are built on top of Pipecat, which is an Open Source framework for stitching together various services like text-to-speech (TTS), LLMs, and speech-to-text (STT). Daily Bots provides a growing set of pre-configured “bot profiles” which setup common combinations of these services for you.

For each type of service, Daily Bots supports a predefined set of providers, outlined below. For the current set of bots, you will configure one LLM provider and one TTS provider with the option of using an integrated one or bringing your own API key. Optionally, you can configure a STT provider in order to specify a non-English language. For integrated providers, you can still bring your own API key, but if one is not provided, Daily uses its own and passes on the cost to you.

Check out our pricing page for more information. Or check out our BYO Keys page for more information on how to bring your own API key.

Speech-to-Text

We support Deepgram as an integrated speech-to-text provider and two additional speech-to-text providers that work by providing your own API key. See Bringing Your Own API Key for more information.

ProviderConfiguration StringIntegratedSupported Models
Deepgram"deepgram"YesAny listed here
AssemblyAI"assemblyai"NoN/A
Gladia"gladia"NoN/A

LLM

We support a limited number of integrated LLM providers along with an expanded list of LLM providers that work with your own API key. See Bringing Your Own API Key for more information. The following list will be expanded as we add more providers in the near future.

ProviderConfiguration StringIntegratedSupported Models
Anthropic"anthropic"Yesclaude-3-5-sonnet-20240620
claude-3-5-sonnet-20241022
claude-3-5-sonnet-latest
claude-3-5-haiku-20241022
claude-3-5-haiku-latest
Together AI"together"Yes"meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo"
"meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo"
"meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo"
"meta-llama/Llama-3.2-3B-Instruct-Turbo"
"meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo"
"meta-llama/Llama-3.2-90B-Instruct-Turbo"
Google Gemini"gemini"NoAny listed here
Grok"grok"NoAny listed here
Groq"groq"NoAny listed here
OpenAI"openai"NoAny listed here
Custom LLM"custom_llm"NoRun any OpenAI spec compliant LLM. Learn more.

Text-to-Speech

We support Cartesia, Deepgram, and ElevenLabs as integrated text-to-speech providers and a number of other text-to-speech providers that work by providing your own API key. See Bringing Your Own API Key for more information. We’re adding new services all the time. Contact us to request a new service.

ProviderConfiguration StringIntegratedSupported Models
Cartesia"cartesia"Yes"sonic-english"
"sonic-multilingual"
Deepgram"deepgram_tts"YesAny listed here
ElevenLabs"elevenlabs"Yeseleven_turbo_v2_5
eleven_turbo_v2
Azure"azure_tts"NoAny listed here
AWS"aws_tts"NoAny listed here
Google"google_tts"NoAny listed here
OpenAI"openai_tts"No"tts-1"
"tts-1-hd"
PlayHT"playht"NoPlayHT3.0-mini
PlayHT2.0-turbo
PlayHT2.0

Speech-to-Speech

Currently we support one speech-to-speech provider, OpenAI Realtime. This list will be expanded in the near future.

ProviderConfiguration StringIntegratedSupported Models
OpenAI"openai_realtime"NoN/A

Transport

Daily is the only supported transport provider. Visit Daily’s service page to learn more about available service options.