deepgram

package

v0.0.6 Latest Latest Go to latest Published: Apr 23, 2026 License: MIT Imports: 17 Imported by: 0

Details

Valid go.mod file
Redistributable license
Tagged version
Stable version
Learn more about best practices

Repository

github.com/cavos-io/rtp-agent

Links

Open Source Insights

Documentation ¶

Index ¶

func IsNova3Model(model string) bool
type DeepgramOption
- func WithKeyterms(keyterms []string) DeepgramOption
- func WithKeywords(keywords []Keyword) DeepgramOption
type DeepgramSTT
- func NewDeepgramSTT(apiKey string, model string, opts ...DeepgramOption) *DeepgramSTT
type DeepgramTTS
- func NewDeepgramTTS(apiKey string, model string) *DeepgramTTS
type Keyword

Constants ¶

This section is empty.

Variables ¶

This section is empty.

Functions ¶

func IsNova3Model ¶ added in v0.0.6

func IsNova3Model(model string) bool

IsNova3Model returns true if the model string indicates a nova-3 or newer model. This is exported for use by application-layer logic to determine keyword parameters.

Types ¶

type DeepgramOption ¶ added in v0.0.6

type DeepgramOption func(*DeepgramSTT)

DeepgramOption is a functional option for configuring DeepgramSTT.

func WithKeyterms ¶ added in v0.0.6

func WithKeyterms(keyterms []string) DeepgramOption

WithKeyterms sets the keyterms (for nova-3 and newer models). Keyterms are passed as keyword strings only (no boost weight) in the Deepgram API.

func WithKeywords ¶ added in v0.0.6

func WithKeywords(keywords []Keyword) DeepgramOption

WithKeywords sets the keywords with boost weights (for nova-2 and older models). Keywords are passed as "keyword:boostInt" format in the Deepgram API.

type DeepgramSTT ¶

type DeepgramSTT struct {
	// contains filtered or unexported fields
}

func NewDeepgramSTT ¶

func NewDeepgramSTT(apiKey string, model string, opts ...DeepgramOption) *DeepgramSTT

func (*DeepgramSTT) Capabilities ¶

func (s *DeepgramSTT) Capabilities() stt.STTCapabilities

func (*DeepgramSTT) Label ¶

func (s *DeepgramSTT) Label() string

func (*DeepgramSTT) Recognize ¶

func (s *DeepgramSTT) Recognize(ctx context.Context, frames []*model.AudioFrame, languageStr string) (*stt.SpeechEvent, error)

func (*DeepgramSTT) Stream ¶

func (s *DeepgramSTT) Stream(ctx context.Context, languageStr string) (stt.RecognizeStream, error)

type DeepgramTTS ¶

type DeepgramTTS struct {
	// contains filtered or unexported fields
}

func NewDeepgramTTS ¶

func NewDeepgramTTS(apiKey string, model string) *DeepgramTTS

func (*DeepgramTTS) Capabilities ¶

func (t *DeepgramTTS) Capabilities() tts.TTSCapabilities

func (*DeepgramTTS) Label ¶

func (t *DeepgramTTS) Label() string

func (*DeepgramTTS) NumChannels ¶

func (t *DeepgramTTS) NumChannels() int

func (*DeepgramTTS) SampleRate ¶

func (t *DeepgramTTS) SampleRate() int

func (*DeepgramTTS) Stream ¶

func (t *DeepgramTTS) Stream(ctx context.Context) (tts.SynthesizeStream, error)

func (*DeepgramTTS) Synthesize ¶

func (t *DeepgramTTS) Synthesize(ctx context.Context, text string) (tts.ChunkedStream, error)

type Keyword ¶ added in v0.0.6

type Keyword struct {
	Keyword string  `json:"keyword"`
	Boost   float64 `json:"boost"`
}

Keyword represents a keyword with a boost weight for Deepgram STT recognition. This is the canonical definition shared between rtp-agent and agent-worker layer.

Source Files ¶

View all Source files

?	: This menu
/	: Search site
f or F	: Jump to
y or Y	: Canonical URL