From 23579d4741609a3273bc38a76463367319f4d653 Mon Sep 17 00:00:00 2001
From: Paul Gauthier <aider@paulg.org>
Date: Mon, 14 Aug 2023 14:11:03 -0300
Subject: [PATCH] Refactor voice recording and transcription to allow for
 specifying the language.

---
 aider/commands.py | 9 +++++++--
 aider/voice.py    | 4 ++--
 2 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/aider/commands.py b/aider/commands.py
index b562ea4e5..f35eb5d49 100644
--- a/aider/commands.py
+++ b/aider/commands.py
@@ -14,9 +14,14 @@ from .dump import dump  # noqa: F401
 
 
 class Commands:
-    def __init__(self, io, coder):
+    def __init__(self, io, coder, voice_language=None):
         self.io = io
         self.coder = coder
+
+        if voice_language == 'auto':
+            voice_language=None
+
+        self.voice_language = voice_language
         self.tokenizer = tiktoken.encoding_for_model(coder.main_model.name)
 
     def is_command(self, inp):
@@ -460,7 +465,7 @@ class Commands:
         history = "\n".join(history)
         dump(history)
 
-        text = v.record_and_transcribe(history)
+        text = v.record_and_transcribe(history, self.voice_language)
         if text:
             self.io.add_to_input_history(text)
             print()
diff --git a/aider/voice.py b/aider/voice.py
index d8b2e92cb..462693ab9 100644
--- a/aider/voice.py
+++ b/aider/voice.py
@@ -60,7 +60,7 @@ class Voice:
         except KeyboardInterrupt:
             return
 
-    def raw_record_and_transcribe(self, history):
+    def raw_record_and_transcribe(self, history, language):
         self.q = queue.Queue()
 
         filename = tempfile.mktemp(suffix=".wav")
@@ -77,7 +77,7 @@ class Voice:
                 file.write(self.q.get())
 
         with open(filename, "rb") as fh:
-            transcript = openai.Audio.transcribe("whisper-1", fh, prompt=history)
+            transcript = openai.Audio.transcribe("whisper-1", fh, prompt=history, language=language)
 
         text = transcript["text"]
         return text