feat(clients): google cloud voice: generic client, TTS, STT, Gemini audio, live voice page

src/google: GoogleClient over molo::clients (key in x-goog-api-key, never in
URLs), Service base, TextToSpeech, SpeechToText, Generative (replies, audio
in, audio out), Recognizer/Synthesizer seams with cloud, gemini and vertex
backends, VoicePipeline with bounded history. Example: /voice live
conversation page, /voice/{say,hear,turn,reset}, /google/status. Keys load
from env, .env.json.local, .env.json. Paid routes are excluded from locust.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
This commit is contained in:
2026-10-24 04:05:41 +02:00
co-authored by Claude Sonnet 5.5
parent dd773d3015
commit 837de58136
54 changed files with 3086 additions and 21 deletions
+19
View File
@@ -0,0 +1,19 @@
#pragma once
// Cloud Speech-to-Text: speech:recognize (synchronous, up to ~1 minute).
#include <string>
#include "google/service.h"
#include "google/voice_io.h"
namespace molo::google {
class SpeechToText : public Service, public Recognizer {
public:
explicit SpeechToText(std::shared_ptr<GoogleClient> client);
Transcript recognize(const RecognitionRequest& request) const override;
};
} // namespace molo::google