deepgram-go-text-to-speech

Using Deepgram Text-to-Speech from the Go SDK

When to use this product

Use this skill for pkg/client/speak work:

• file or stream synthesis over REST
• low-latency synthesis over WebSockets
• callback-based or channel-based audio playback pipelines

Use a different skill when:

• you need STT (deepgram-go-speech-to-text)
• you need live voice-agent orchestration (deepgram-go-voice-agent)
• you need repo workflow guidance (deepgram-go-maintaining-sdk)

Authentication

Set DEEPGRAM_API_KEY before creating Speak clients.

export DEEPGRAM_API_KEY="your_api_key"

Use the repo's env-backed client defaults instead of embedding secrets in code.

Quick start

REST synthesis to file:

package main

import (
	"context"
	"log"

	api "github.com/deepgram/deepgram-go-sdk/v3/pkg/api/speak/v1/rest"
	speak "github.com/deepgram/deepgram-go-sdk/v3/pkg/client/speak"
	interfaces "github.com/deepgram/deepgram-go-sdk/v3/pkg/client/interfaces"
)

func main() {
	if err := run(); err != nil {
		log.Fatal(err)
	}
}

func run() error {
	ctx := context.Background()

	client := speak.NewRESTWithDefaults()
	dg := api.New(client)

	if _, err := dg.ToSave(
		ctx,
		"hello.wav",
		"Hello from the Deepgram Go SDK.",
		&interfaces.SpeakOptions{Model: "aura-2-thalia-en"},
	); err != nil {
		return err
	}

	return nil
}

Streaming synthesis with callbacks or channels:

package main

import (
	"context"
	"fmt"
	"log"

	speakws "github.com/deepgram/deepgram-go-sdk/v3/pkg/api/speak/v1/websocket"
	speak "github.com/deepgram/deepgram-go-sdk/v3/pkg/client/speak"
	interfaces "github.com/deepgram/deepgram-go-sdk/v3/pkg/client/interfaces"
)

func main() {
	if err := run(); err != nil {
		log.Fatal(err)
	}
}

func run() error {
	ctx := context.Background()
	handler := speakws.NewDefaultChanHandler()

	conn, err := speak.NewWSUsingChanWithDefaults(
		ctx,
		&interfaces.WSSpeakOptions{Model: "aura-2-thalia-en"},
		handler,
	)
	if err != nil {
		return err
	}
	defer conn.Stop()

	if ok := conn.Connect(); !ok {
		return fmt.Errorf("connect failed")
	}

	conn.Start()

	if err := conn.SpeakWithText("Streaming TTS from Go."); err != nil {
		return err
	}

	// The handler receives binary audio and flow-control events.
	if err := conn.Flush(); err != nil {
		return err
	}

	return nil
}

Key parameters

• interfaces.SpeakOptions
  • typical fields: Model, Encoding, Container, SampleRate
• interfaces.WSSpeakOptions
  • typical fields: Model, streaming audio format settings
• REST methods
  • via pkg/api/speak/v1/rest: api.New(client).ToStream, ToFile, ToSave
• WS methods
  • SpeakWithText, Speak, Flush, Clear
  • Reset is deprecated; use Clear instead
• constructors
  • speak.NewRESTWithDefaults() / speak.NewREST(...)
  • speak.NewWSUsingCallback...
  • speak.NewWSUsingChan...

API reference (layered)

1. In-repo reference
   • README.md
   • docs.go
   • pkg/client/speak/client.go
   • pkg/client/speak/v1/rest/client.go
   • pkg/client/speak/v1/websocket/client_callback.go
   • pkg/client/speak/v1/websocket/client_channel.go
   • pkg/client/interfaces/v1/types-speak.go
2. OpenAPI
   • https://developers.deepgram.com/openapi.yaml
3. AsyncAPI
   • https://developers.deepgram.com/asyncapi.yaml
4. Context7
   • /llmstxt/developers_deepgram_llms_txt
5. Product docs
   • https://developers.deepgram.com/reference/text-to-speech/speak-request
   • https://developers.deepgram.com/reference/text-to-speech/speak-streaming
   • https://developers.deepgram.com/docs/tts-models

Gotchas

1. REST and WebSocket Speak clients use different option structs.
2. REST methods live on pkg/api/speak/v1/rest; build them with api.New(client).
3. WebSocket flows use a handler passed at construction, Connect() returns bool, and shutdown is Stop().
4. Keep the playback or message-consumer goroutine running while audio frames arrive.
5. Follow the examples for file handling and output format selection instead of guessing encodings.

Example files in this repo

• examples/text-to-speech/rest/file/hello-world/main.go
• examples/text-to-speech/websocket/simple_channel/main.go
• examples/text-to-speech/websocket/simple_callback/main.go

Central product skills

For cross-language Deepgram product knowledge — the consolidated API reference, documentation finder, focused runnable recipes, third-party integration examples, and MCP setup — install the central skills:

npx skills add deepgram/skills

This SDK ships language-idiomatic code skills; deepgram/skills ships cross-language product knowledge (see api, docs, recipes, examples, starters, setup-mcp).