From 966050a7b453b5e443ac8ed986e48e850b86f66d Mon Sep 17 00:00:00 2001
From: talksik <patel.arjun50@gmail.com>
Date: Sat, 26 Jul 2025 19:30:03 -0700
Subject: [PATCH] transcribe raw audio

---
 ai-conversation/audio_capture.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

diff --git a/ai-conversation/audio_capture.c b/ai-conversation/audio_capture.c
index f88a027..4a5ee3f 100644
--- a/ai-conversation/audio_capture.c
+++ b/ai-conversation/audio_capture.c
@@ -33,6 +33,7 @@ size_t read_callback(void *ptr, size_t size, size_t nmemb, void *userdata) {
 // Returns string that caller must free
 char* find_transcript(char* json_response)
 {
+  printf("Parsing response: %s \n", json_response);
   char* start = strstr(json_response, "\"transcript\":\"");
   if (!start) return NULL;
 
@@ -61,6 +62,8 @@ size_t write_callback(void *contents, size_t size, size_t nmemb, void *userp) {
     response->size += total_size;
     response->data[response->size] = '\0'; // Null terminate
     
+    printf("read an extra %d bytes of data\n", (int)total_size);
+    
     return total_size;
 }
 
@@ -88,9 +91,8 @@ char* transcribe(FILE* audio_file) {
     if (curl) {
         // Set headers for Deepgram API
         headers = curl_slist_append(headers, "Authorization: Token YOUR_DEEPGRAM_API_KEY");
-        headers = curl_slist_append(headers, "Content-Type: audio/*");
 
-        curl_easy_setopt(curl, CURLOPT_URL, "https://api.deepgram.com/v1/listen");
+        curl_easy_setopt(curl, CURLOPT_URL, "https://api.deepgram.com/v1/listen?encoding=linear16&sample_rate=48000&channels=1");
         curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
         curl_easy_setopt(curl, CURLOPT_POST, 1L);
         curl_easy_setopt(curl, CURLOPT_READFUNCTION, read_callback);
@@ -108,6 +110,8 @@ char* transcribe(FILE* audio_file) {
             return NULL;
         }
 
+        printf("done performing curl\n");
+
         curl_slist_free_all(headers);
         curl_easy_cleanup(curl);
     }
@@ -214,9 +218,8 @@ int main() {
       perror("Error opening file");
       return 1;
     }
+    printf("======READING FRAMES for 5 seconds========\n");
     while (snd_pcm_readi(pcm_handle, (void *)buf, period_size) > 0 && periods_read < periods_to_read) {
-        printf("======READING FRAMES========\n");
-
         // Because each frame is 2 bytes (format/bit rate), cast to int16
         // int16_t *samples = (int16_t *)buf;
         // for (int i = 0; i < period_size; i++) {
@@ -224,10 +227,9 @@ int main() {
         // }
 
         fwrite(buf, sizeof(int16_t), period_size, file);
-
-        printf("\n");
         periods_read++;
     }
+    printf("done capturing audio\n");
 
     fclose(file);
     
@@ -238,10 +240,14 @@ int main() {
         if (transcript) {
             printf("Transcript: %s\n", transcript);
             free(transcript);
+        } else {
+            fprintf(stderr, "unable to get transcript");
         }
         fclose(file);
+    } else {
+      fprintf(stderr, "unable to open file for reading\n");
     }
-    
+
     free((void *)buf);
     snd_pcm_close(pcm_handle);
     snd_pcm_hw_params_free(params);