create full flow with stt, anthropic, tts, and playback

This organizes some components like deepgram into own module, and also
allows gets the full flow to work every time we run the program.

ai-conversation/deepgram_client.h

@@ -0,0 +1,21 @@
+#ifndef DEEPGRAMCLIENT_H
+#define DEEPGRAMCLIENT_H
+
+#include <curl/curl.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include "utils.h"
+
+/*
+ * @param audio_file: must already be opened
+ * @returns char *: transcript string result. Call must free.
+ */
+char *dg_transcribe(FILE *audio_file);
+
+/*
+*  @returns response data. Caller must free the struct AND the data within it.
+*  TODO: should return custom struct with details about encoding, sample rate, etc.
+*/
+struct curl_response_data *dg_text_to_speech(char *text);
+
+#endif // DEEPGRAMCLIENT_H