natural-tts 0.3.0

High-level bindings to a variety of text-to-speech libraries.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119

use std::path::PathBuf;

use super::{AudioHandler, NaturalModelTrait, Spec, SynthesizedAudio};
use crate::utils::{play_audio, save_wav};
use msedge_tts::{
    tts::{client::connect, SpeechConfig as OtherConfig},
    voice::{get_voices_list, Voice},
};

#[derive(Clone, Debug)]
pub struct MSEdgeModel {
    config: SpeechConfig,
}

impl MSEdgeModel {
    pub fn new_from_voice(voice: Voice) -> Self {
        Self {
            config: SpeechConfig::from(&voice),
        }
    }

    pub fn new(config: SpeechConfig) -> Self {
        Self { config }
    }
}

impl Default for MSEdgeModel {
    fn default() -> Self {
        let voice = get_voices_list().unwrap();
        Self::new(SpeechConfig::from(voice.first().unwrap()))
    }
}

impl NaturalModelTrait for MSEdgeModel {
    type SynthesizeType = f32;
    fn start(
        &mut self,
        message: String,
        path: &PathBuf,
    ) -> Result<AudioHandler, Box<dyn std::error::Error>> {
        let synthesized = Self::synthesize(self, message, path)?;

        let rate = match self.config.rate {
            x if x <= 0 => 16000,
            x => x,
        };

        Ok(AudioHandler::from(match synthesized.spec {
            Spec::Wav(x) => play_audio(synthesized.data, x.sample_rate),
            _ => play_audio(synthesized.data, rate as u32),
        }?))
    }

    fn save(&mut self, message: String, path: &PathBuf) -> Result<(), Box<dyn std::error::Error>> {
        let synthesized = Self::synthesize(self, message, path)?;

        let rate = match self.config.rate {
            x if x <= 0 => 16000,
            x => x,
        };

        let _ = save_wav(&synthesized.data, path, rate as u32);
        Ok(())
    }

    fn synthesize(
        &mut self,
        message: String,
        _path: &PathBuf,
    ) -> Result<super::SynthesizedAudio<Self::SynthesizeType>, Box<dyn std::error::Error>> {
        let mut tts = connect().unwrap();
        let audio = tts.synthesize(message.as_str(), &self.config.as_msedge())?;
        Ok(SynthesizedAudio::new(
            audio.audio_bytes.iter().map(|x| *x as f32).collect(),
            Spec::Synthesized(audio.audio_format, audio.audio_metadata),
            None,
        ))
    }
}

#[derive(Debug, Clone)]
pub struct SpeechConfig {
    pub voice_name: String,
    pub audio_format: String,
    pub pitch: i32,
    pub rate: i32,
    pub volume: i32,
}

impl SpeechConfig {
    pub fn as_msedge(&self) -> OtherConfig {
        OtherConfig {
            voice_name: self.voice_name.clone(),
            audio_format: self.audio_format.clone(),
            pitch: self.pitch,
            rate: self.rate,
            volume: self.volume,
        }
    }
}

impl From<&msedge_tts::tts::SpeechConfig> for SpeechConfig {
    fn from(config: &msedge_tts::tts::SpeechConfig) -> Self {
        Self {
            voice_name: config.voice_name.clone(),
            audio_format: config.audio_format.clone(),
            pitch: config.pitch,
            rate: config.rate,
            volume: config.volume,
        }
    }
}

impl From<&msedge_tts::voice::Voice> for SpeechConfig {
    fn from(voice: &msedge_tts::voice::Voice) -> Self {
        let mscfg = OtherConfig::from(voice);
        Self::from(&mscfg)
    }
}