Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

support custom tts streaming #475

Closed
wants to merge 2 commits into from
Closed

support custom tts streaming #475

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
67d1156
support custom tts streaming
xquanluu Jan 7, 2025
6396e3b
wip
xquanluu Jan 8, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions src/api/types.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -434,6 +434,10 @@ export interface SpeechCredential {
deepgram_tts_uri: null | string;
deepgram_stt_use_tls: number;
speechmatics_stt_uri: null | string;
use_audio_resampling: number;
use_base64_encoding: number;
tts_streaming_sample_rate: number;
use_tls: number;
}

export interface Alert {
Expand Down
99 changes: 98 additions & 1 deletion src/containers/internal/views/speech-services/form.tsx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -184,6 +184,15 @@ export const SpeechServiceForm = ({ credential }: SpeechServiceFormProps) => {
AWS_CREDENTIAL_ACCESS_KEY,
);
const [roleArn, setRoleArn] = useState("");
const [useAudioResampling, setUseAudioResampling] = useState(false);
const [tmpUseAudioResampling, setTmpUseAudioResampling] = useState(false);
const [useBase64Encoding, setUseBase64Encoding] = useState(false);
const [tmpUseBase64Encoding, setTmpUseBase64Encoding] = useState(false);
const [useTls, setUseTls] = useState(false);
const [tmpUseTls, setTmpUseTls] = useState(false);
const [ttsStreamingSampleRate, setTtsStreamingSampleRate] = useState("8000");
const [tmpTtsStreamingSampleRate, setTmpTtsStreamingSampleRate] =
useState("8000");

const handleFile = (file: File) => {
const handleError = () => {
Expand Down Expand Up @@ -381,6 +390,10 @@ export const SpeechServiceForm = ({ credential }: SpeechServiceFormProps) => {
custom_tts_url: customVendorTtsUrl || null,
custom_stt_url: customVendorSttUrl || null,
auth_token: customVendorAuthToken || null,
use_audio_resampling: useAudioResampling ? 1 : 0,
use_base64_encoding: useBase64Encoding ? 1 : 0,
use_tls: useTls ? 1 : 0,
tts_streaming_sample_rate: Number(ttsStreamingSampleRate || "8000"),
}),
...(vendor === VENDOR_NUANCE && {
client_id: clientId || null,
Expand Down Expand Up @@ -756,6 +769,26 @@ export const SpeechServiceForm = ({ credential }: SpeechServiceFormProps) => {
);
setSpeechmaticsEndpoint(credential.data.speechmatics_stt_uri);
}
setUseAudioResampling(
(credential?.data?.use_audio_resampling || 0) > 0 ? true : false,
);
setTmpUseAudioResampling(
(credential?.data?.use_audio_resampling || 0) > 0 ? true : false,
);
setUseBase64Encoding(
(credential?.data?.use_base64_encoding || 0) > 0 ? true : false,
);
setTmpUseBase64Encoding(
(credential?.data?.use_base64_encoding || 0) > 0 ? true : false,
);
if (credential?.data?.tts_streaming_sample_rate) {
setTtsStreamingSampleRate(`${credential.data.tts_streaming_sample_rate}`);
setTmpTtsStreamingSampleRate(
`${credential.data.tts_streaming_sample_rate}`,
);
}
setUseTls((credential?.data?.use_tls || 0) > 0 ? true : false);
setTmpUseTls((credential?.data?.use_tls || 0) > 0 ? true : false);
}, [credential]);

const updateCustomVoices = (
Expand Down Expand Up @@ -914,14 +947,25 @@ export const SpeechServiceForm = ({ credential }: SpeechServiceFormProps) => {
setTtsCheck(e.target.checked);
if (!e.target.checked) {
setTmpCustomVendorTtsUrl(customVendorTtsUrl);
setTmpUseAudioResampling(useAudioResampling);
setTmpUseTls(useTls);
setTmpUseBase64Encoding(useBase64Encoding);
setCustomVendorTtsUrl("");
setUseAudioResampling(false);
setUseTls(false);
setUseBase64Encoding(false);
setTmpTtsStreamingSampleRate(ttsStreamingSampleRate);
} else {
setCustomVendorTtsUrl(tmpCustomVendorTtsUrl);
setUseAudioResampling(tmpUseAudioResampling);
setUseTls(tmpUseTls);
setUseBase64Encoding(tmpUseBase64Encoding);
setTtsStreamingSampleRate(tmpTtsStreamingSampleRate);
}
}}
>
<label htmlFor="custom_vendor_use_for_tts">
TTS HTTP URL<span>*</span>
TTS URL<span>*</span>
</label>
<input
id="custom_vendor_use_for_tts"
Expand All @@ -934,6 +978,59 @@ export const SpeechServiceForm = ({ credential }: SpeechServiceFormProps) => {
setCustomVendorTtsUrl(e.target.value);
}}
/>
<label htmlFor="use_audio_resampling" className="chk">
<input
id="use_audio_resampling"
name="use_audio_resampling"
type="checkbox"
onChange={(e) => setUseAudioResampling(e.target.checked)}
defaultChecked={useAudioResampling}
/>
<div>Tts streaming use audio resampling</div>
</label>

{useAudioResampling && (
<>
{" "}
<label htmlFor="custom_vendor_tts_streaming_sample_rate">
Tts streaming audio sample rate
{useAudioResampling ? <span>*</span> : ""}
</label>
<input
id="custom_vendor_tts_streaming_sample_rate"
type="number"
name="custom_vendor_tts_streaming_sample_rate"
required={useAudioResampling}
value={ttsStreamingSampleRate}
disabled={!useAudioResampling}
onChange={(e) => {
setTtsStreamingSampleRate(e.target.value);
}}
/>
</>
)}

<label htmlFor="use_base64_encoding" className="chk">
<input
id="use_base64_encoding"
name="use_base64_encoding"
type="checkbox"
onChange={(e) => setUseBase64Encoding(e.target.checked)}
defaultChecked={useBase64Encoding}
/>
<div>Tts streaming use base64 encoding</div>
</label>

<label htmlFor="use_tls" className="chk">
<input
id="use_tls"
name="use_tls"
type="checkbox"
onChange={(e) => setUseTls(e.target.checked)}
defaultChecked={useTls}
/>
<div>Tts streaming use TLS</div>
</label>
</Checkzone>

<Checkzone
Expand Down
Loading