Add text-to-speech support. #13
7 changed files with 116 additions and 122 deletions
|
|
@ -63,21 +63,18 @@ func initPhysicalDevices(config *config.ConfigParser) map[string]*evdev.InputDev
|
|||
func main() {
|
||||
// parse command-line
|
||||
var configFlag string
|
||||
var ttsVoiceFlag string
|
||||
var ttsFlag bool
|
||||
flag.BoolVarP(&logger.IsDebugMode, "debug", "d", false, "Output very verbose debug messages.")
|
||||
flag.StringVarP(&configFlag, "config", "c", "~/.config/joyful", "Directory to read configuration from.")
|
||||
addTTSFlags(&ttsFlag, &ttsVoiceFlag)
|
||||
ttsOps := addTTSFlags()
|
||||
flag.Parse()
|
||||
|
||||
// parse configs
|
||||
configDir := getConfigDir(configFlag)
|
||||
config := readConfig(configDir)
|
||||
|
||||
tts, err := newTTS(ttsFlag, ttsVoiceFlag)
|
||||
// initialize TTS
|
||||
tts, err := newTTS(ttsOps)
|
||||
logger.LogIfError(err, "Failed to initialize TTS")
|
||||
if tts != nil {
|
||||
defer tts.Cleanup()
|
||||
}
|
||||
|
||||
// Initialize virtual devices with event buffers
|
||||
vBuffersByName, vBuffersByDevice := initVirtualBuffers(config)
|
||||
|
|
@ -91,6 +88,12 @@ func main() {
|
|||
// initialize the mode variable
|
||||
mode := config.GetModes()[0]
|
||||
|
||||
// initialize TTS phrases for modes
|
||||
for _, m := range config.GetModes() {
|
||||
tts.AddMessage(m)
|
||||
logger.LogDebugf("Added TTS message '%s'", m)
|
||||
}
|
||||
|
||||
fmt.Println("Joyful Running! Press Ctrl+C to quit. Press Enter to reload rules.")
|
||||
if len(config.GetModes()) > 1 {
|
||||
logger.Logf("Initial mode set to '%s'", mode)
|
||||
|
|
|
|||
|
|
@ -1,65 +1,51 @@
|
|||
//go:build !notts
|
||||
|
||||
package main
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"io"
|
||||
"os"
|
||||
"fmt"
|
||||
"os/exec"
|
||||
"strconv"
|
||||
"time"
|
||||
|
||||
"git.annabunches.net/annabunches/joyful/internal/logger"
|
||||
"github.com/amitybell/piper"
|
||||
asset "github.com/amitybell/piper-asset"
|
||||
alan "github.com/amitybell/piper-voice-alan"
|
||||
jenny "github.com/amitybell/piper-voice-jenny"
|
||||
"github.com/ebitengine/oto/v3"
|
||||
flag "github.com/spf13/pflag"
|
||||
)
|
||||
|
||||
type TTSOptions struct {
|
||||
Disabled bool
|
||||
Voice string
|
||||
Volume int
|
||||
Pitch int
|
||||
Range int
|
||||
Speed int
|
||||
}
|
||||
|
||||
type TTS struct {
|
||||
piper.TTS
|
||||
dataDir string
|
||||
options *TTSOptions
|
||||
otoCtx *oto.Context
|
||||
phrases map[string][]byte
|
||||
}
|
||||
|
||||
const (
|
||||
playbackCheckIntervalMs = 250
|
||||
playbackSeekOffsetBytes = 1024
|
||||
playbackCheckIntervalMs = 100
|
||||
)
|
||||
|
||||
func addTTSFlags(ttsFlag *bool, ttsVoiceFlag *string) {
|
||||
flag.BoolVar(ttsFlag, "notts", false, "Disable text-to-speech on mode change.")
|
||||
flag.StringVar(ttsVoiceFlag, "voice", "alan", "Which voice to use for TTS; must be 'alan' or 'jenny'")
|
||||
// TODO: make most of this configurable via file
|
||||
func addTTSFlags() *TTSOptions {
|
||||
ops := &TTSOptions{}
|
||||
|
||||
flag.BoolVar(&ops.Disabled, "no-tts", false, "Disable text-to-speech.")
|
||||
flag.StringVar(&ops.Voice, "tts-voice", "en", "Which voice to use for TTS; see 'espeak --voices' for a full list of options.")
|
||||
flag.IntVar(&ops.Volume, "tts-volume", 100, "Text to speech volume")
|
||||
flag.IntVar(&ops.Pitch, "tts-pitch", 50, "Text to speech volume")
|
||||
flag.IntVar(&ops.Range, "tts-range", 50, "Text to speech volume")
|
||||
flag.IntVar(&ops.Range, "tts-speed", 175, "Text to speech speaking speed (in words per minute)")
|
||||
|
||||
return ops
|
||||
}
|
||||
|
||||
func newTTS(disable bool, voice string) (*TTS, error) {
|
||||
if disable {
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
dataDir, err := os.MkdirTemp("", "joyful-piper.")
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
var ass asset.Asset
|
||||
switch voice {
|
||||
case "jenny":
|
||||
ass = jenny.Asset
|
||||
case "alan":
|
||||
ass = alan.Asset
|
||||
default:
|
||||
ass = alan.Asset
|
||||
}
|
||||
|
||||
pTTS, err := piper.NewEmbedded(dataDir, ass)
|
||||
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
func makeOtoContext() (*oto.Context, error) {
|
||||
op := &oto.NewContextOptions{
|
||||
SampleRate: 22050,
|
||||
ChannelCount: 1,
|
||||
|
|
@ -72,35 +58,78 @@ func newTTS(disable bool, voice string) (*TTS, error) {
|
|||
}
|
||||
<-readyChan // wait for initialization
|
||||
|
||||
return otoCtx, nil
|
||||
}
|
||||
|
||||
func newTTS(ops *TTSOptions) (*TTS, error) {
|
||||
if ops.Disabled {
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
context, err := makeOtoContext()
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
return &TTS{
|
||||
TTS: *pTTS,
|
||||
dataDir: dataDir,
|
||||
otoCtx: otoCtx,
|
||||
options: ops,
|
||||
otoCtx: context,
|
||||
phrases: make(map[string][]byte),
|
||||
}, nil
|
||||
}
|
||||
|
||||
// "Say" generates TTS audio and plays it in a go routine
|
||||
func (t *TTS) Say(msg string) {
|
||||
go func() {
|
||||
wav, err := t.Synthesize(msg)
|
||||
func (t *TTS) AddMessage(msg string) {
|
||||
// TODO: need to get lots of input validation in here
|
||||
// We execute `espeak-ng` directly because extant libraries produce terrible output
|
||||
// compared to the command-line utility. This also gives us a chance to
|
||||
cmd := exec.Command(
|
||||
"espeak-ng", "--stdout",
|
||||
"-v", t.options.Voice,
|
||||
"-a", strconv.Itoa(t.options.Volume),
|
||||
"-p", strconv.Itoa(t.options.Pitch),
|
||||
"-P", strconv.Itoa(t.options.Range),
|
||||
"-s", strconv.Itoa(t.options.Speed),
|
||||
msg,
|
||||
)
|
||||
|
||||
if err != nil {
|
||||
logger.LogError(err, "")
|
||||
return
|
||||
wavData, err := cmd.Output()
|
||||
if err != nil {
|
||||
logger.LogError(err, "Failed to create TTS data")
|
||||
return
|
||||
}
|
||||
|
||||
t.phrases[msg] = wavData
|
||||
}
|
||||
|
||||
// "Say" generates TTS audio and plays it in a go routine
|
||||
func (t *TTS) Say(msg string) error {
|
||||
if _, ok := t.phrases[msg]; !ok {
|
||||
return fmt.Errorf("tried to play non-buffered phrase '%s'", msg)
|
||||
}
|
||||
|
||||
go func(buf []byte) {
|
||||
buffer := bytes.NewBuffer(buf)
|
||||
player := t.otoCtx.NewPlayer(buffer)
|
||||
|
||||
volume := 0.0
|
||||
player.SetVolume(volume)
|
||||
player.Play()
|
||||
|
||||
// Gradually ramp up the volume to avoid harsh clicks
|
||||
for player.Volume() < 1.0 {
|
||||
volume += 0.01
|
||||
if volume > 1.0 {
|
||||
volume = 1.0
|
||||
}
|
||||
|
||||
player.SetVolume(volume)
|
||||
time.Sleep(1 * time.Millisecond)
|
||||
}
|
||||
|
||||
wavReader := bytes.NewReader(wav)
|
||||
player := t.otoCtx.NewPlayer(wavReader)
|
||||
// We seek some bytes into the generated audio because there's a click
|
||||
// and a long delay at the beginning of the data.
|
||||
player.Seek(playbackSeekOffsetBytes, io.SeekStart)
|
||||
player.Play()
|
||||
for player.IsPlaying() {
|
||||
time.Sleep(playbackCheckIntervalMs * time.Millisecond)
|
||||
}
|
||||
}()
|
||||
}
|
||||
}(t.phrases[msg])
|
||||
|
||||
func (t *TTS) Cleanup() {
|
||||
os.RemoveAll(t.dataDir)
|
||||
return nil
|
||||
}
|
||||
|
|
|
|||
|
|
@ -1,16 +0,0 @@
|
|||
//go:build notts
|
||||
|
||||
package main
|
||||
|
||||
type Speaker interface {
|
||||
Say(string)
|
||||
Cleanup()
|
||||
}
|
||||
|
||||
func newTTS(_ bool, _ string) (Speaker, error) {
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
func addTTSFlags(ttsFlag *bool, ttsVoiceFlag *string) {
|
||||
return
|
||||
}
|
||||
8
go.mod
8
go.mod
|
|
@ -3,10 +3,6 @@ module git.annabunches.net/annabunches/joyful
|
|||
go 1.24.4
|
||||
|
||||
require (
|
||||
github.com/amitybell/piper v0.0.0-20250621082041-2bb74e3a4a55
|
||||
github.com/amitybell/piper-asset v0.0.0-20231030194325-d36a29e3b1fd
|
||||
github.com/amitybell/piper-voice-alan v0.0.0-20231118093148-059963c24dbd
|
||||
github.com/amitybell/piper-voice-jenny v0.0.0-20231118093224-dcf0d49e46b7
|
||||
github.com/ebitengine/oto/v3 v3.3.3
|
||||
github.com/goccy/go-yaml v1.18.0
|
||||
github.com/holoplot/go-evdev v0.0.0-20240306072622-217e18f17db1
|
||||
|
|
@ -17,12 +13,8 @@ require (
|
|||
)
|
||||
|
||||
require (
|
||||
github.com/adrg/xdg v0.5.3 // indirect
|
||||
github.com/amitybell/piper-bin-linux v0.0.0-20250621082830-f5d5d85fa076 // indirect
|
||||
github.com/amitybell/piper-bin-windows v0.0.0-20231118093113-cc2cef2f6b74 // indirect
|
||||
github.com/davecgh/go-spew v1.1.1 // indirect
|
||||
github.com/ebitengine/purego v0.8.4 // indirect
|
||||
github.com/klauspost/compress v1.18.0 // indirect
|
||||
github.com/pmezard/go-difflib v1.0.0 // indirect
|
||||
github.com/stretchr/objx v0.5.2 // indirect
|
||||
golang.org/x/sys v0.34.0 // indirect
|
||||
|
|
|
|||
18
go.sum
18
go.sum
|
|
@ -1,19 +1,3 @@
|
|||
github.com/adrg/xdg v0.5.3 h1:xRnxJXne7+oWDatRhR1JLnvuccuIeCoBu2rtuLqQB78=
|
||||
github.com/adrg/xdg v0.5.3/go.mod h1:nlTsY+NNiCBGCK2tpm09vRqfVzrc2fLmXGpBLF0zlTQ=
|
||||
github.com/amitybell/piper v0.0.0-20250621082041-2bb74e3a4a55 h1:8MKEDgDBbYKAphlRUcAEHT1Uam7xBjA5E/SmGHhNH10=
|
||||
github.com/amitybell/piper v0.0.0-20250621082041-2bb74e3a4a55/go.mod h1:y0aDZdCM3erPmpX+rDGoF0O2ZdCqZvAxNjYUPrK/O7U=
|
||||
github.com/amitybell/piper-asset v0.0.0-20231030194325-d36a29e3b1fd h1:4MLHn2cCVhzhPLlPO6946h1S0yk3o7Ry1831DEa5EcE=
|
||||
github.com/amitybell/piper-asset v0.0.0-20231030194325-d36a29e3b1fd/go.mod h1:MiDKnt4NenfcrsVxYAxQW0nu4zjFYQPjGzzLB5MvOz8=
|
||||
github.com/amitybell/piper-bin-linux v0.0.0-20250621082830-f5d5d85fa076 h1:aST7iEpuMr507piwgx0WNDezW6ycWIE+ejtnXXaMgI0=
|
||||
github.com/amitybell/piper-bin-linux v0.0.0-20250621082830-f5d5d85fa076/go.mod h1:dVR33O0l/AFgQNmZfywfgNZ6qlpCKPhLnn9UpeMMLdM=
|
||||
github.com/amitybell/piper-bin-windows v0.0.0-20231118093113-cc2cef2f6b74 h1:T5hXX0Z2JaE5gtZ7LScjG0r0BmDk0+FWlzyZ2b1nboo=
|
||||
github.com/amitybell/piper-bin-windows v0.0.0-20231118093113-cc2cef2f6b74/go.mod h1:5Ea0Pc0QdO8FeriIXcqZtHViM2fi589jtFubrjaAk6w=
|
||||
github.com/amitybell/piper-voice-alan v0.0.0-20231118093148-059963c24dbd h1:DsXuiWSHsbBkVNL7cBAdXD95kNwrE0Ck05OasSeUZ4g=
|
||||
github.com/amitybell/piper-voice-alan v0.0.0-20231118093148-059963c24dbd/go.mod h1:5ghO6mSctWNXfDoh3r46HQEMIcPr5DqE5TMYfp5hskY=
|
||||
github.com/amitybell/piper-voice-jenny v0.0.0-20231030195502-2afb5ebf3c45 h1:V/HZAQuprvdo0xXToxAuTLSwD3YrqRpDZLVBOOD+2aE=
|
||||
github.com/amitybell/piper-voice-jenny v0.0.0-20231030195502-2afb5ebf3c45/go.mod h1:eKG2Bo69QGTVKKKKApafZr+4v4zk40jYNijh0s8/PzU=
|
||||
github.com/amitybell/piper-voice-jenny v0.0.0-20231118093224-dcf0d49e46b7 h1:GMYJcgP1OKBMBuQfP7r0aRk4PS0AaviHVTERtdt/e/o=
|
||||
github.com/amitybell/piper-voice-jenny v0.0.0-20231118093224-dcf0d49e46b7/go.mod h1:eKG2Bo69QGTVKKKKApafZr+4v4zk40jYNijh0s8/PzU=
|
||||
github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
|
||||
github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
|
||||
github.com/ebitengine/oto/v3 v3.3.3 h1:m6RV69OqoXYSWCDsHXN9rc07aDuDstGHtait7HXSM7g=
|
||||
|
|
@ -26,8 +10,6 @@ github.com/holoplot/go-evdev v0.0.0-20240306072622-217e18f17db1 h1:92OsBIf5KB1Ta
|
|||
github.com/holoplot/go-evdev v0.0.0-20240306072622-217e18f17db1/go.mod h1:iHAf8OIncO2gcQ8XOjS7CMJ2aPbX2Bs0wl5pZyanEqk=
|
||||
github.com/jonboulle/clockwork v0.5.0 h1:Hyh9A8u51kptdkR+cqRpT1EebBwTn1oK9YfGYbdFz6I=
|
||||
github.com/jonboulle/clockwork v0.5.0/go.mod h1:3mZlmanh0g2NDKO5TWZVJAfofYk64M7XN3SzBPjZF60=
|
||||
github.com/klauspost/compress v1.18.0 h1:c/Cqfb0r+Yi+JtIEq73FWXVkRonBlf0CRNYc8Zttxdo=
|
||||
github.com/klauspost/compress v1.18.0/go.mod h1:2Pp+KzxcywXVXMr50+X0Q/Lsb43OQHYWRCY2AiWywWQ=
|
||||
github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
|
||||
github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
|
||||
github.com/spf13/pflag v1.0.7 h1:vN6T9TfwStFPFM5XzjsvmzZkLuaLX+HS+0SeFLRgU6M=
|
||||
|
|
|
|||
|
|
@ -5,6 +5,8 @@ import (
|
|||
"os"
|
||||
)
|
||||
|
||||
var IsDebugMode = false
|
||||
|
||||
func Log(msg string) {
|
||||
fmt.Println(msg)
|
||||
}
|
||||
|
|
@ -13,6 +15,12 @@ func Logf(msg string, params ...interface{}) {
|
|||
fmt.Printf(msg+"\n", params...)
|
||||
}
|
||||
|
||||
func LogDebugf(msg string, params ...interface{}) {
|
||||
if IsDebugMode {
|
||||
fmt.Printf("DEBUG: %s\n", fmt.Sprintf(msg, params...))
|
||||
}
|
||||
}
|
||||
|
||||
func LogError(err error, msg string) {
|
||||
if msg == "" {
|
||||
fmt.Printf("%s\n", err.Error())
|
||||
|
|
|
|||
20
readme.md
20
readme.md
|
|
@ -20,6 +20,7 @@ Joyful is ideal for Linux gamers who enjoy space and flight sims and miss the fe
|
|||
* Axis -> Relative Axis mapping, for converting a joystick axis to mouse movement and scrollwheel events.
|
||||
* Configure per-rule configurable deadzones for axes, with multiple ways to specify deadzones.
|
||||
* Define multiple modes with per-mode behavior.
|
||||
* Text-to-speech engine that announces the current mode when it changes.
|
||||
|
||||
### Possible Future Features
|
||||
|
||||
|
|
@ -47,23 +48,18 @@ Pressing `<enter>` in the running terminal window will reload the `rules` sectio
|
|||
|
||||
## Build & Install
|
||||
|
||||
To build joyful, first use your distribution's package manager to install `go` and `alsa-lib` (this may be `libasound2-dev` or `libasound2-devel` depending on your distribution), then run:
|
||||
To build joyful, first use your distribution's package manager to install the following packages:
|
||||
* `go`
|
||||
* `alsa-lib` - this may be `libasound2-dev` or `libasound2-devel` depending on your distribution
|
||||
* `espeak-ng` - if you want text-to-speech to announce mode changes
|
||||
|
||||
Then, run:
|
||||
|
||||
```
|
||||
go build -o build/ ./...
|
||||
```
|
||||
|
||||
Next, copy the binaries in the `build/` directory to somewhere in your `$PATH`. (details depend on your setup, but typically somewhere like `/usr/local/bin` or `~/bin`)
|
||||
|
||||
### Machine Learning Disclosure
|
||||
|
||||
Joyful's text-to-speech support is dependent on [Piper](https://github.com/rhasspy/piper), which uses an offline Machine Learning (ML) model for speech synthesis. The project authors are extremely skeptical of ML/AI technologies in general, but consider speech synthesis, especially offline/local speech synthesis, to be one of the most defensible use cases for it. Since it is very difficult to find text-to-speech systems that don't use ML under the hood (especially that have extant golang wrappers or bindings), this is considered a necessary tradeoff.
|
||||
|
||||
However, if you don't want any ML running on your system, you can optionally choose to skip TTS support at compile-time by building with this command:
|
||||
|
||||
```
|
||||
go build -o build -tags notts ./...
|
||||
```
|
||||
Finally, copy the files in the `build/` directory to somewhere in your `$PATH`. (details depend on your setup, but typically somewhere like `/usr/local/bin` or `~/bin`)
|
||||
|
||||
## Technical details
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue