diff --git a/esphome/components/rtttl/rtttl.cpp b/esphome/components/rtttl/rtttl.cpp
index ab95067f45..4ccfc539ea 100644
--- a/esphome/components/rtttl/rtttl.cpp
+++ b/esphome/components/rtttl/rtttl.cpp
@@ -8,18 +8,26 @@ namespace esphome::rtttl {
 
 static const char *const TAG = "rtttl";
 
-// These values can also be found as constants in the Tone library (Tone.h)
-static const uint16_t NOTES[] = {0,    262,  277,  294,  311,  330,  349,  370,  392,  415,  440,  466,  494,
-                                 523,  554,  587,  622,  659,  698,  740,  784,  831,  880,  932,  988,  1047,
-                                 1109, 1175, 1245, 1319, 1397, 1480, 1568, 1661, 1760, 1865, 1976, 2093, 2217,
-                                 2349, 2489, 2637, 2794, 2960, 3136, 3322, 3520, 3729, 3951};
+static constexpr uint8_t SONG_NAME_LENGTH_LIMIT = 64;
+static constexpr uint8_t SEMITONES_IN_OCTAVE = 12;
 
-#if defined(USE_OUTPUT) || defined(USE_SPEAKER)
-static const uint32_t DOUBLE_NOTE_GAP_MS = 10;
-#endif  // USE_OUTPUT || USE_SPEAKER
+static constexpr uint8_t MIN_OCTAVE = 4;
+static constexpr uint8_t MAX_OCTAVE = 7;
+
+static constexpr uint8_t DEFAULT_BPM = 63;  // Default beats per minute
+
+// These values can also be found as constants in the Tone library (Tone.h)
+static constexpr uint16_t NOTES[] = {0,    262,  277,  294,  311,  330,  349,  370,  392,  415,  440,  466,  494,
+                                     523,  554,  587,  622,  659,  698,  740,  784,  831,  880,  932,  988,  1047,
+                                     1109, 1175, 1245, 1319, 1397, 1480, 1568, 1661, 1760, 1865, 1976, 2093, 2217,
+                                     2349, 2489, 2637, 2794, 2960, 3136, 3322, 3520, 3729, 3951};
+static constexpr uint8_t NOTES_COUNT = static_cast<uint8_t>(sizeof(NOTES) / sizeof(NOTES[0]));
+
+static constexpr uint8_t REPEATING_NOTE_GAP_MS = 10;
 
 #ifdef USE_SPEAKER
-static const size_t SAMPLE_BUFFER_SIZE = 2048;
+static constexpr uint16_t SAMPLE_BUFFER_SIZE = 2048;
+static constexpr uint16_t SAMPLE_RATE = 16000;
 
 struct SpeakerSample {
   int8_t left{0};
@@ -27,7 +35,7 @@ struct SpeakerSample {
 };
 
 inline double deg2rad(double degrees) {
-  static const double PI_ON_180 = 4.0 * atan(1.0) / 180.0;
+  static constexpr double PI_ON_180 = M_PI / 180.0;
   return degrees * PI_ON_180;
 }
 #endif  // USE_SPEAKER
@@ -85,7 +93,7 @@ void Rtttl::loop() {
   }
 
 #ifdef USE_OUTPUT
-  if (this->output_ != nullptr && millis() - this->last_note_ < this->note_duration_) {
+  if (this->output_ != nullptr && millis() - this->last_note_start_time_ < this->note_duration_) {
     return;
   }
 #endif  // USE_OUTPUT
@@ -113,36 +121,34 @@ void Rtttl::loop() {
     }
     if (this->samples_sent_ != this->samples_count_) {
       SpeakerSample sample[SAMPLE_BUFFER_SIZE + 2];
-      int x = 0;
+      uint16_t sample_index = 0;
       double rem = 0.0;
 
       while (true) {
-        // Try and send out the remainder of the existing note, one per loop()
-
-        if (this->samples_per_wave_ != 0 && this->samples_sent_ >= this->samples_gap_) {  // Play note//
+        // Try and send out the remainder of the existing note, one per `loop()`
+        if (this->samples_per_wave_ != 0 && this->samples_sent_ >= this->samples_gap_) {  // Play note
           rem = ((this->samples_sent_ << 10) % this->samples_per_wave_) * (360.0 / this->samples_per_wave_);
 
-          int16_t val = (127 * this->gain_) * sin(deg2rad(rem));  // 16bit = 49152
-
-          sample[x].left = val;
-          sample[x].right = val;
+          int8_t val = (127 * this->gain_) * sin(deg2rad(rem));
 
+          sample[sample_index].left = val;
+          sample[sample_index].right = val;
         } else {
-          sample[x].left = 0;
-          sample[x].right = 0;
+          sample[sample_index].left = 0;
+          sample[sample_index].right = 0;
         }
 
-        if (static_cast<size_t>(x) >= SAMPLE_BUFFER_SIZE || this->samples_sent_ >= this->samples_count_) {
+        if (sample_index >= SAMPLE_BUFFER_SIZE || this->samples_sent_ >= this->samples_count_) {
           break;
         }
         this->samples_sent_++;
-        x++;
+        sample_index++;
       }
-      if (x > 0) {
-        size_t bytes_to_send = x * sizeof(SpeakerSample);
+      if (sample_index > 0) {
+        size_t bytes_to_send = sample_index * sizeof(SpeakerSample);
         size_t send = this->speaker_->play((uint8_t *) (&sample), bytes_to_send);
         if (send != bytes_to_send) {
-          this->samples_sent_ -= (x - (send / sizeof(SpeakerSample)));
+          this->samples_sent_ -= (sample_index - (send / sizeof(SpeakerSample)));
         }
         return;
       }
@@ -155,83 +161,84 @@ void Rtttl::loop() {
     return;
   }
 
-  // align to note: most rtttl's out there does not add and space after the ',' separator but just in case...
+  // Align to note: most rtttl's out there does not add any space after the ',' separator but just in case
   while (this->rtttl_[this->position_] == ',' || this->rtttl_[this->position_] == ' ') {
     this->position_++;
   }
 
-  // first, get note duration, if available
-  uint8_t num = this->get_integer_();
+  // First, get note duration, if available
+  uint8_t note_denominator = this->get_integer_();
 
-  if (num) {
-    this->note_duration_ = this->wholenote_ / num;
+  if (note_denominator) {
+    this->note_duration_ = this->wholenote_duration_ / note_denominator;
   } else {
-    this->note_duration_ =
-        this->wholenote_ / this->default_duration_;  // we will need to check if we are a dotted note after
+    // We will need to check if we are a dotted note after
+    this->note_duration_ = this->wholenote_duration_ / this->default_note_denominator_;
   }
 
-  uint8_t note = note_index_from_char(this->rtttl_[this->position_]);
+  uint8_t note_index_in_octave = note_index_from_char(this->rtttl_[this->position_]);
 
   this->position_++;
 
-  // now, get optional '#' sharp
+  // Now, get optional '#' sharp
   if (this->rtttl_[this->position_] == '#') {
-    note++;
+    note_index_in_octave++;
     this->position_++;
   }
 
-  // now, get scale
+  // Now, get scale
   uint8_t scale = this->get_integer_();
   if (scale == 0) {
     scale = this->default_octave_;
   }
 
-  if (scale < 4 || scale > 7) {
-    ESP_LOGE(TAG, "Octave must be between 4 and 7 (it is %d)", scale);
+  if (scale < MIN_OCTAVE || scale > MAX_OCTAVE) {
+    ESP_LOGE(TAG, "Octave must be between %d and %d (it is %d)", MIN_OCTAVE, MAX_OCTAVE, scale);
     this->finish_();
     return;
   }
 
-  // now, get optional '.' dotted note
+  // Now, get optional '.' dotted note
   if (this->rtttl_[this->position_] == '.') {
-    this->note_duration_ += this->note_duration_ / 2;
+    this->note_duration_ += this->note_duration_ / 2;  // Duration +50%
     this->position_++;
   }
 
-  // Now play the note
   bool need_note_gap = false;
-  if (note) {
-    auto note_index = (scale - 4) * 12 + note;
-    if (note_index < 0 || note_index >= (int) (sizeof(NOTES) / sizeof(NOTES[0]))) {
-      ESP_LOGE(TAG, "Note out of range (note: %d, scale: %d, index: %d, max: %d)", note, scale, note_index,
-               (int) (sizeof(NOTES) / sizeof(NOTES[0])));
+
+  // Now play the note
+  if (note_index_in_octave == 0) {
+    this->output_freq_ = 0;
+    ESP_LOGVV(TAG, "Waiting: %dms", this->note_duration_);
+  } else {
+    uint8_t note_index = (scale - MIN_OCTAVE) * SEMITONES_IN_OCTAVE + note_index_in_octave;
+    if (note_index >= NOTES_COUNT) {
+      ESP_LOGE(TAG, "Note out of range (note: %d, scale: %d, index: %d, max: %d)", note_index_in_octave, scale,
+               note_index, NOTES_COUNT);
       this->finish_();
       return;
     }
-    auto freq = NOTES[note_index];
+    uint16_t freq = NOTES[note_index];
     need_note_gap = freq == this->output_freq_;
 
     // Add small silence gap between same note
     this->output_freq_ = freq;
 
-    ESP_LOGVV(TAG, "playing note: %d for %dms", note, this->note_duration_);
-  } else {
-    ESP_LOGVV(TAG, "waiting: %dms", this->note_duration_);
-    this->output_freq_ = 0;
+    ESP_LOGVV(TAG, "Playing note: %d for %dms", note_index_in_octave, this->note_duration_);
   }
 
 #ifdef USE_OUTPUT
   if (this->output_ != nullptr) {
-    if (need_note_gap && this->note_duration_ > DOUBLE_NOTE_GAP_MS) {
+    if (this->output_freq_ == 0) {
       this->output_->set_level(0.0);
-      delay(DOUBLE_NOTE_GAP_MS);
-      this->note_duration_ -= DOUBLE_NOTE_GAP_MS;
-    }
-    if (this->output_freq_ != 0) {
+    } else {
+      if (need_note_gap && this->note_duration_ > REPEATING_NOTE_GAP_MS) {
+        this->output_->set_level(0.0);
+        delay(REPEATING_NOTE_GAP_MS);
+        this->note_duration_ -= REPEATING_NOTE_GAP_MS;
+      }
       this->output_->update_frequency(this->output_freq_);
       this->output_->set_level(this->gain_);
-    } else {
-      this->output_->set_level(0.0);
     }
   }
 #endif  // USE_OUTPUT
@@ -241,28 +248,26 @@ void Rtttl::loop() {
     this->samples_sent_ = 0;
     this->samples_gap_ = 0;
     this->samples_per_wave_ = 0;
-    this->samples_count_ = (this->sample_rate_ * this->note_duration_) / 1000;
+    this->samples_count_ = (SAMPLE_RATE * this->note_duration_) / 1000;
     if (need_note_gap) {
-      this->samples_gap_ = (this->sample_rate_ * DOUBLE_NOTE_GAP_MS) / 1000;
+      this->samples_gap_ = (SAMPLE_RATE * REPEATING_NOTE_GAP_MS) / 1000;
     }
     if (this->output_freq_ != 0) {
-      // make sure there is enough samples to add a full last sinus.
-
-      uint16_t samples_wish = this->samples_count_;
-      this->samples_per_wave_ = (this->sample_rate_ << 10) / this->output_freq_;
+      // Make sure there is enough samples to add a full last sinus.
+      uint32_t samples_wish = this->samples_count_;
+      this->samples_per_wave_ = (SAMPLE_RATE << 10) / this->output_freq_;
 
       uint16_t division = ((this->samples_count_ << 10) / this->samples_per_wave_) + 1;
 
-      this->samples_count_ = (division * this->samples_per_wave_);
-      this->samples_count_ = this->samples_count_ >> 10;
-      ESP_LOGVV(TAG, "- Calc play time: wish: %d gets: %d (div: %d spw: %d)", samples_wish, this->samples_count_,
-                division, this->samples_per_wave_);
+      this->samples_count_ = (division * this->samples_per_wave_) >> 10;
+      ESP_LOGVV(TAG, "Calc play time: wish: %" PRIu32 " gets: %" PRIu32 " (div: %d spw: %" PRIu32 ")", samples_wish,
+                this->samples_count_, division, this->samples_per_wave_);
     }
     // Convert from frequency in Hz to high and low samples in fixed point
   }
 #endif  // USE_SPEAKER
 
-  this->last_note_ = millis();
+  this->last_note_start_time_ = millis();
 }
 
 void Rtttl::play(std::string rtttl) {
@@ -275,25 +280,28 @@ void Rtttl::play(std::string rtttl) {
 
   this->rtttl_ = std::move(rtttl);
 
-  this->default_duration_ = 4;
-  this->default_octave_ = 6;
+  this->default_note_denominator_ = DEFAULT_NOTE_DENOMINATOR;
+  this->default_octave_ = DEFAULT_OCTAVE;
   this->note_duration_ = 0;
 
-  int bpm = 63;
-  uint16_t num;
+  uint16_t bpm = DEFAULT_BPM;
+  uint16_t num;  // Used for: default note-denominator, default octave, BPM
 
   // Get name
   this->position_ = this->rtttl_.find(':');
 
-  // it's somewhat documented to be up to 10 characters but let's be a bit flexible here
-  if (this->position_ == std::string::npos || this->position_ > 15) {
+  if (this->position_ == std::string::npos) {
     ESP_LOGE(TAG, "Unable to determine name; missing ':'");
     return;
   }
-
+  if (this->position_ >= SONG_NAME_LENGTH_LIMIT) {
+    ESP_LOGE(TAG, "Name is too long: length=%u, limit=%u", static_cast<unsigned>(this->position_),
+             static_cast<unsigned>(SONG_NAME_LENGTH_LIMIT));
+    return;
+  }
   ESP_LOGD(TAG, "Playing song %.*s", (int) this->position_, this->rtttl_.c_str());
 
-  // get default duration
+  // Get default duration
   this->position_ = this->rtttl_.find("d=", this->position_);
   if (this->position_ == std::string::npos) {
     ESP_LOGE(TAG, "Missing 'd='");
@@ -301,11 +309,14 @@ void Rtttl::play(std::string rtttl) {
   }
   this->position_ += 2;
   num = this->get_integer_();
-  if (num > 0) {
-    this->default_duration_ = num;
+  if (num == 1 || num == 2 || num == 4 || num == 8 || num == 16 || num == 32) {
+    this->default_note_denominator_ = num;
+  } else {
+    ESP_LOGE(TAG, "Invalid default duration: %d", num);
+    return;
   }
 
-  // get default octave
+  // Get default octave
   this->position_ = this->rtttl_.find("o=", this->position_);
   if (this->position_ == std::string::npos) {
     ESP_LOGE(TAG, "Missing 'o=");
@@ -313,11 +324,14 @@ void Rtttl::play(std::string rtttl) {
   }
   this->position_ += 2;
   num = this->get_integer_();
-  if (num >= 3 && num <= 7) {
+  if (num >= MIN_OCTAVE && num <= MAX_OCTAVE) {
     this->default_octave_ = num;
+  } else {
+    ESP_LOGE(TAG, "Invalid default octave: %d", num);
+    return;
   }
 
-  // get BPM
+  // Get BPM
   this->position_ = this->rtttl_.find("b=", this->position_);
   if (this->position_ == std::string::npos) {
     ESP_LOGE(TAG, "Missing b=");
@@ -325,8 +339,11 @@ void Rtttl::play(std::string rtttl) {
   }
   this->position_ += 2;
   num = this->get_integer_();
-  if (num != 0) {
+  if (num >= 4) {  // Below 4 is not realistic and would cause a integer overflow
     bpm = num;
+  } else {
+    ESP_LOGE(TAG, "Invalid BPM: %d", num);
+    return;
   }
 
   this->position_ = this->rtttl_.find(':', this->position_);
@@ -337,10 +354,10 @@ void Rtttl::play(std::string rtttl) {
   this->position_++;
 
   // BPM usually expresses the number of quarter notes per minute
-  this->wholenote_ = 60 * 1000L * 4 / bpm;  // this is the time for whole note (in milliseconds)
+  this->wholenote_duration_ = 60 * 1000L * 4 / bpm;  // This is the time for whole note (in milliseconds)
 
   this->output_freq_ = 0;
-  this->last_note_ = millis();
+  this->last_note_start_time_ = millis();
   this->note_duration_ = 1;
 
 #ifdef USE_OUTPUT
diff --git a/esphome/components/rtttl/rtttl.h b/esphome/components/rtttl/rtttl.h
index 4d4a652c51..e37cccae9e 100644
--- a/esphome/components/rtttl/rtttl.h
+++ b/esphome/components/rtttl/rtttl.h
@@ -13,6 +13,10 @@
 
 namespace esphome::rtttl {
 
+inline constexpr uint8_t DEFAULT_NOTE_DENOMINATOR = 4;  // Default note-denominator (quarter note)
+inline constexpr uint8_t DEFAULT_OCTAVE =
+    6;  // Default octave for a note (see: `MIN_OCTAVE`, `MAX_OCTAVE` in `rtttl.cpp`)
+
 enum class State : uint8_t {
   STOPPED = 0,
   INIT,
@@ -67,19 +71,18 @@ class Rtttl : public Component {
   std::string rtttl_{""};
   /// The current position in the RTTTL string.
   size_t position_{0};
-  /// The duration of a whole note in milliseconds.
-  uint16_t wholenote_;
   /// The default duration of a note (e.g. 4 for a quarter note).
-  uint16_t default_duration_;
+  uint8_t default_note_denominator_{DEFAULT_NOTE_DENOMINATOR};
   /// The default octave for a note.
-  uint16_t default_octave_;
-  /// The time the last note was started.
-  uint32_t last_note_;
+  uint8_t default_octave_{DEFAULT_OCTAVE};
   /// The duration of the current note in milliseconds.
-  uint16_t note_duration_;
-
+  uint16_t note_duration_{0};
+  /// The duration of a whole note in milliseconds.
+  uint16_t wholenote_duration_;
+  /// The time in milliseconds since microcontroller boot when the last note was started.
+  uint32_t last_note_start_time_;
   /// The frequency of the current note in Hz.
-  uint32_t output_freq_;
+  uint32_t output_freq_{0};
   /// The gain of the output.
   float gain_{0.6f};
   /// The current state of the RTTTL player.
@@ -93,16 +96,14 @@ class Rtttl : public Component {
 #ifdef USE_SPEAKER
   /// The speaker to write the sound to.
   speaker::Speaker *speaker_{nullptr};
-  /// The sample rate of the speaker.
-  int sample_rate_{16000};
   /// The number of samples for one full cycle of a note's waveform, in Q10 fixed-point format.
-  int samples_per_wave_{0};
+  uint32_t samples_per_wave_{0};
   /// The number of samples sent.
-  int samples_sent_{0};
+  uint32_t samples_sent_{0};
   /// The total number of samples to send.
-  int samples_count_{0};
+  uint32_t samples_count_{0};
   /// The number of samples for the gap between notes.
-  int samples_gap_{0};
+  uint32_t samples_gap_{0};
 #endif  // USE_SPEAKER
 
   /// The callback to call when playback is finished.