diff --git a/src/audio/mixin_mixout/mixin_mixout.c b/src/audio/mixin_mixout/mixin_mixout.c
index 9c1abe30386c..576b46bb31f7 100644
--- a/src/audio/mixin_mixout/mixin_mixout.c
+++ b/src/audio/mixin_mixout/mixin_mixout.c
@@ -75,6 +75,7 @@ struct mixin_sink_config {
 /* mixin component private data */
 struct mixin_data {
 	mix_func mix;
+	mix_func gain_mix;
 	struct mixin_sink_config sink_config[MIXIN_MAX_SINKS];
 };
 
@@ -205,8 +206,13 @@ static int mix(struct comp_dev *dev, const struct mixin_data *mixin_data,
 
 	sink_config = &mixin_data->sink_config[sink_index];
 
-	mixin_data->mix(sink, start_sample, mixed_samples,
+	if (sink_config->gain == IPC4_MIXIN_UNITY_GAIN) {
+		mixin_data->mix(sink, start_sample, mixed_samples,
 			source, sample_count, sink_config->gain);
+	} else {
+		mixin_data->gain_mix(sink, start_sample, mixed_samples,
+			     source, sample_count, sink_config->gain);
+	}
 
 	return 0;
 }
@@ -555,6 +561,7 @@ static int mixin_reset(struct processing_module *mod)
 	comp_dbg(dev, "mixin_reset()");
 
 	mixin_data->mix = NULL;
+	mixin_data->gain_mix = NULL;
 
 	return 0;
 }
@@ -685,15 +692,15 @@ static int mixin_prepare(struct processing_module *mod,
 	case SOF_IPC_FRAME_S16_LE:
 	case SOF_IPC_FRAME_S24_4LE:
 	case SOF_IPC_FRAME_S32_LE:
-		md->mix = mixin_get_processing_function(fmt);
+		mixin_get_processing_functions(fmt, &md->mix, &md->gain_mix);
 		break;
 	default:
 		comp_err(dev, "unsupported data format %d", fmt);
 		return -EINVAL;
 	}
 
-	if (!md->mix) {
-		comp_err(dev, "have not found the suitable processing function");
+	if (!md->mix || !md->gain_mix) {
+		comp_err(dev, "have not found suitable processing functions");
 		return -EINVAL;
 	}
 
diff --git a/src/audio/mixin_mixout/mixin_mixout.h b/src/audio/mixin_mixout/mixin_mixout.h
index 085187e0e743..d7519ca80fd8 100644
--- a/src/audio/mixin_mixout/mixin_mixout.h
+++ b/src/audio/mixin_mixout/mixin_mixout.h
@@ -110,7 +110,8 @@ typedef void (*mix_func)(struct cir_buf_ptr *sink, int32_t start_sample,
  */
 struct mix_func_map {
 	uint16_t frame_fmt;	/* frame format */
-	mix_func func;		/* mixin processing function */
+	mix_func mix;		/* faster mixing func without gain support */
+	mix_func gain_mix;	/* slower mixing func with gain support */
 };
 
 extern const struct mix_func_map mix_func_map[];
@@ -119,17 +120,23 @@ extern const size_t mix_count;
  * \brief Retrievies mixin processing function.
  * \param[in] fmt  stream PCM frame format
  */
-static inline mix_func mixin_get_processing_function(int fmt)
+static inline bool mixin_get_processing_functions(int fmt, mix_func *mix, mix_func *gain_mix)
 {
 	int i;
 
+	*mix = NULL;
+	*gain_mix = NULL;
+
 	/* map mixin processing function for source and sink buffers */
 	for (i = 0; i < mix_count; i++) {
-		if (fmt == mix_func_map[i].frame_fmt)
-			return mix_func_map[i].func;
+		if (fmt == mix_func_map[i].frame_fmt) {
+			*mix = mix_func_map[i].mix;
+			*gain_mix = mix_func_map[i].gain_mix;
+			return true;
+		}
 	}
 
-	return NULL;
+	return false;
 }
 
 #endif	/* __SOF_IPC4_MIXIN_MIXOUT_H__ */
diff --git a/src/audio/mixin_mixout/mixin_mixout_generic.c b/src/audio/mixin_mixout/mixin_mixout_generic.c
index 60a9f8211b05..69ba5f02a4bd 100644
--- a/src/audio/mixin_mixout/mixin_mixout_generic.c
+++ b/src/audio/mixin_mixout/mixin_mixout_generic.c
@@ -54,6 +54,54 @@ static void mix_s16(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixe
 		src += n;
 	}
 }
+
+static void mix_s16_gain(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixed_samples,
+			 const struct cir_buf_ptr *source,
+			 int32_t sample_count, uint16_t gain)
+{
+	int32_t samples_to_mix, samples_to_copy, left_samples;
+	int32_t n, nmax, i;
+
+	/* cir_buf_wrap() is required and is done below in a loop */
+	int16_t *dst = (int16_t *)sink->ptr + start_sample;
+	int16_t *src = source->ptr;
+
+	assert(mixed_samples >= start_sample);
+	samples_to_mix = mixed_samples - start_sample;
+	samples_to_mix = MIN(samples_to_mix, sample_count);
+	samples_to_copy = sample_count - samples_to_mix;
+
+	for (left_samples = samples_to_mix; left_samples > 0; left_samples -= n) {
+		src = cir_buf_wrap(src, source->buf_start, source->buf_end);
+		dst = cir_buf_wrap(dst, sink->buf_start, sink->buf_end);
+		/* calculate the remaining samples*/
+		nmax = (int16_t *)source->buf_end - src;
+		n = MIN(left_samples, nmax);
+		nmax = (int16_t *)sink->buf_end - dst;
+		n = MIN(n, nmax);
+		for (i = 0; i < n; i++) {
+			*dst = sat_int16((int32_t)*dst +
+				q_mults_16x16(*src, gain, IPC4_MIXIN_GAIN_SHIFT));
+			src++;
+			dst++;
+		}
+	}
+
+	for (left_samples = samples_to_copy; left_samples > 0; left_samples -= n) {
+		src = cir_buf_wrap(src, source->buf_start, source->buf_end);
+		dst = cir_buf_wrap(dst, sink->buf_start, sink->buf_end);
+		nmax = (int16_t *)source->buf_end - src;
+		n = MIN(left_samples, nmax);
+		nmax = (int16_t *)sink->buf_end - dst;
+		n = MIN(n, nmax);
+
+		for (i = 0; i < n; i++) {
+			*dst = q_mults_16x16(*src, gain, IPC4_MIXIN_GAIN_SHIFT);
+			src++;
+			dst++;
+		}
+	}
+}
 #endif	/* CONFIG_FORMAT_S16LE */
 
 #if CONFIG_FORMAT_S24LE
@@ -99,6 +147,52 @@ static void mix_s24(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixe
 	}
 }
 
+static void mix_s24_gain(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixed_samples,
+			 const struct cir_buf_ptr *source,
+			 int32_t sample_count, uint16_t gain)
+{
+	int32_t samples_to_mix, samples_to_copy, left_samples;
+	int32_t n, nmax, i;
+	/* cir_buf_wrap() is required and is done below in a loop */
+	int32_t *dst = (int32_t *)sink->ptr + start_sample;
+	int32_t *src = source->ptr;
+
+	assert(mixed_samples >= start_sample);
+	samples_to_mix = mixed_samples - start_sample;
+	samples_to_mix = MIN(samples_to_mix, sample_count);
+	samples_to_copy = sample_count - samples_to_mix;
+
+	for (left_samples = samples_to_mix; left_samples > 0; left_samples -= n) {
+		src = cir_buf_wrap(src, source->buf_start, source->buf_end);
+		dst = cir_buf_wrap(dst, sink->buf_start, sink->buf_end);
+		/* calculate the remaining samples*/
+		nmax = (int32_t *)source->buf_end - src;
+		n = MIN(left_samples, nmax);
+		nmax = (int32_t *)sink->buf_end - dst;
+		n = MIN(n, nmax);
+		for (i = 0; i < n; i++) {
+			*dst = sat_int24(sign_extend_s24(*dst) +
+				(int32_t)q_mults_32x32(sign_extend_s24(*src),
+						       gain, IPC4_MIXIN_GAIN_SHIFT));
+			src++;
+			dst++;
+		}
+	}
+
+	for (left_samples = samples_to_copy; left_samples > 0; left_samples -= n) {
+		src = cir_buf_wrap(src, source->buf_start, source->buf_end);
+		dst = cir_buf_wrap(dst, sink->buf_start, sink->buf_end);
+		nmax = (int32_t *)source->buf_end - src;
+		n = MIN(left_samples, nmax);
+		nmax = (int32_t *)sink->buf_end - dst;
+		n = MIN(n, nmax);
+		for (i = 0; i < n; i++) {
+			*dst = q_mults_32x32(sign_extend_s24(*src), gain, IPC4_MIXIN_GAIN_SHIFT);
+			src++;
+			dst++;
+		}
+	}
+}
 #endif	/* CONFIG_FORMAT_S24LE */
 
 #if CONFIG_FORMAT_S32LE
@@ -143,17 +237,61 @@ static void mix_s32(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixe
 	}
 }
 
+static void mix_s32_gain(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixed_samples,
+			 const struct cir_buf_ptr *source,
+			 int32_t sample_count, uint16_t gain)
+{
+	int32_t samples_to_mix, samples_to_copy, left_samples;
+	int32_t n, nmax, i;
+	int32_t *dst = (int32_t *)sink->ptr + start_sample;
+	int32_t *src = source->ptr;
+
+	assert(mixed_samples >= start_sample);
+	samples_to_mix = mixed_samples - start_sample;
+	samples_to_mix = MIN(samples_to_mix, sample_count);
+	samples_to_copy = sample_count - samples_to_mix;
+
+	for (left_samples = samples_to_mix; left_samples > 0; left_samples -= n) {
+		src = cir_buf_wrap(src, source->buf_start, source->buf_end);
+		dst = cir_buf_wrap(dst, sink->buf_start, sink->buf_end);
+		/* calculate the remaining samples*/
+		nmax = (int32_t *)source->buf_end - src;
+		n = MIN(left_samples, nmax);
+		nmax = (int32_t *)sink->buf_end - dst;
+		n = MIN(n, nmax);
+		for (i = 0; i < n; i++) {
+			*dst = sat_int32((int64_t)*dst +
+				q_mults_32x32(*src, gain, IPC4_MIXIN_GAIN_SHIFT));
+			src++;
+			dst++;
+		}
+	}
+
+	for (left_samples = samples_to_copy; left_samples > 0; left_samples -= n) {
+		src = cir_buf_wrap(src, source->buf_start, source->buf_end);
+		dst = cir_buf_wrap(dst, sink->buf_start, sink->buf_end);
+		nmax = (int32_t *)source->buf_end - src;
+		n = MIN(left_samples, nmax);
+		nmax = (int32_t *)sink->buf_end - dst;
+		n = MIN(n, nmax);
+		for (i = 0; i < n; i++) {
+			*dst = q_mults_32x32(*src, gain, IPC4_MIXIN_GAIN_SHIFT);
+			src++;
+			dst++;
+		}
+	}
+}
 #endif	/* CONFIG_FORMAT_S32LE */
 
 const struct mix_func_map mix_func_map[] = {
 #if CONFIG_FORMAT_S16LE
-	{ SOF_IPC_FRAME_S16_LE, mix_s16 },
+	{ SOF_IPC_FRAME_S16_LE, mix_s16, mix_s16_gain },
 #endif
 #if CONFIG_FORMAT_S24LE
-	{ SOF_IPC_FRAME_S24_4LE, mix_s24 },
+	{ SOF_IPC_FRAME_S24_4LE, mix_s24, mix_s24_gain },
 #endif
 #if CONFIG_FORMAT_S32LE
-	{ SOF_IPC_FRAME_S32_LE, mix_s32 }
+	{ SOF_IPC_FRAME_S32_LE, mix_s32, mix_s32_gain }
 #endif
 };
 
diff --git a/src/audio/mixin_mixout/mixin_mixout_hifi3.c b/src/audio/mixin_mixout/mixin_mixout_hifi3.c
index 568cbed5202b..5f0e786fba57 100644
--- a/src/audio/mixin_mixout/mixin_mixout_hifi3.c
+++ b/src/audio/mixin_mixout/mixin_mixout_hifi3.c
@@ -97,6 +97,120 @@ static void mix_s16(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixe
 		}
 	}
 }
+
+static void mix_s16_gain(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixed_samples,
+			 const struct cir_buf_ptr *source,
+			 int32_t sample_count, uint16_t gain)
+{
+	int samples_to_mix, samples_to_copy, left_samples;
+	int n, nmax, i, m, left;
+	ae_int16x4 in_sample;
+	ae_int16x4 out_sample;
+	ae_int16x4 *in;
+	ae_int16x4 *out;
+	ae_valign inu = AE_ZALIGN64();
+	ae_valign outu1 = AE_ZALIGN64();
+	ae_valign outu2 = AE_ZALIGN64();
+	/* cir_buf_wrap() is required and is done below in a loop */
+	ae_int16 *dst = (ae_int16 *)sink->ptr + start_sample;
+	ae_int16 *src = source->ptr;
+	ae_int16x4 gain_vec;
+	ae_int32x2 tmpl, tmph;
+
+	gain_vec = AE_L16_I((ae_int16 *)&gain, 0);
+
+	assert(mixed_samples >= start_sample);
+	samples_to_mix = AE_MIN_32_signed(mixed_samples - start_sample, sample_count);
+	samples_to_copy = sample_count - samples_to_mix;
+	n = 0;
+
+	for (left_samples = samples_to_mix; left_samples > 0; left_samples -= n) {
+		src = cir_buf_wrap(src + n, source->buf_start, source->buf_end);
+		dst = cir_buf_wrap(dst + n, sink->buf_start, sink->buf_end);
+		/* calculate the remaining samples*/
+		nmax = (ae_int16 *)source->buf_end - src;
+		n = AE_MIN_32_signed(left_samples, nmax);
+		nmax = (ae_int16 *)sink->buf_end - dst;
+		n = AE_MIN_32_signed(n, nmax);
+		in = (ae_int16x4 *)src;
+		out = (ae_int16x4 *)dst;
+		inu = AE_LA64_PP(in);
+		outu1 = AE_LA64_PP(out);
+		m = n >> 2;
+		left = n & 0x03;
+		/* process 4 frames per loop */
+		for (i = 0; i < m; i++) {
+			AE_LA16X4_IP(in_sample, inu, in);
+
+			/* apply gain to in_sample */
+			AE_MUL16X4(tmph, tmpl, in_sample, gain_vec);
+			tmpl = AE_SRAI32(tmpl, IPC4_MIXIN_GAIN_SHIFT);
+			tmph = AE_SRAI32(tmph, IPC4_MIXIN_GAIN_SHIFT);
+			in_sample = AE_CVT16X4(tmph, tmpl);
+
+			AE_LA16X4_IP(out_sample, outu1, out);
+			out--;
+			out_sample = AE_ADD16S(in_sample, out_sample);
+			AE_SA16X4_IP(out_sample, outu2, out);
+		}
+		AE_SA64POS_FP(outu2, out);
+
+		/* process the left samples that less than 4
+		 * one by one to avoid memory access overrun
+		 */
+		for (i = 0; i < left ; i++) {
+			AE_L16_IP(in_sample, (ae_int16 *)in, sizeof(ae_int16));
+
+			AE_MUL16X4(tmph, tmpl, in_sample, gain_vec);
+			tmpl = AE_SRAI32(tmpl, IPC4_MIXIN_GAIN_SHIFT);
+			in_sample = AE_CVT16X4(tmpl, tmpl);
+
+			AE_L16_IP(out_sample, (ae_int16 *)out, 0);
+			out_sample = AE_ADD16S(in_sample, out_sample);
+			AE_S16_0_IP(out_sample, (ae_int16 *)out, sizeof(ae_int16));
+		}
+	}
+
+	for (left_samples = samples_to_copy; left_samples > 0; left_samples -= n) {
+		src = cir_buf_wrap(src + n, source->buf_start, source->buf_end);
+		dst = cir_buf_wrap(dst + n, sink->buf_start, sink->buf_end);
+		/* calculate the remaining samples*/
+		nmax = (ae_int16 *)source->buf_end - src;
+		n = AE_MIN_32_signed(left_samples, nmax);
+		nmax = (ae_int16 *)sink->buf_end - dst;
+		n = AE_MIN_32_signed(n, nmax);
+		in = (ae_int16x4 *)src;
+		out = (ae_int16x4 *)dst;
+		inu = AE_LA64_PP(in);
+		m = n >> 2;
+		left = n & 0x03;
+		/* process 4 frames per loop */
+		for (i = 0; i < m; i++) {
+			AE_LA16X4_IP(in_sample, inu, in);
+
+			AE_MUL16X4(tmph, tmpl, in_sample, gain_vec);
+			tmpl = AE_SRAI32(tmpl, IPC4_MIXIN_GAIN_SHIFT);
+			tmph = AE_SRAI32(tmph, IPC4_MIXIN_GAIN_SHIFT);
+			in_sample = AE_CVT16X4(tmph, tmpl);
+
+			AE_SA16X4_IP(in_sample, outu2, out);
+		}
+		AE_SA64POS_FP(outu2, out);
+
+		/* process the left samples that less than 4
+		 * one by one to avoid memory access overrun
+		 */
+		for (i = 0; i < left ; i++) {
+			AE_L16_IP(in_sample, (ae_int16 *)in, sizeof(ae_int16));
+
+			AE_MUL16X4(tmph, tmpl, in_sample, gain_vec);
+			tmpl = AE_SRAI32(tmpl, IPC4_MIXIN_GAIN_SHIFT);
+			in_sample = AE_CVT16X4(tmpl, tmpl);
+
+			AE_S16_0_IP(in_sample, (ae_int16 *)out, sizeof(ae_int16));
+		}
+	}
+}
 #endif	/* CONFIG_FORMAT_S16LE */
 
 #if CONFIG_FORMAT_S24LE
@@ -180,6 +294,121 @@ static void mix_s24(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixe
 	}
 }
 
+static void mix_s24_gain(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixed_samples,
+			 const struct cir_buf_ptr *source,
+			 int32_t sample_count, uint16_t gain)
+{
+	int samples_to_mix, samples_to_copy, left_samples;
+	int n, nmax, i, m, left;
+	ae_int32x2 in_sample, in_sample32;
+	ae_int32x2 out_sample;
+	ae_int32x2 *in;
+	ae_int32x2 *out;
+	ae_valign inu = AE_ZALIGN64();
+	ae_valign outu1 = AE_ZALIGN64();
+	ae_valign outu2 = AE_ZALIGN64();
+	/* cir_buf_wrap() is required and is done below in a loop */
+	int32_t *dst = (int32_t *)sink->ptr + start_sample;
+	int32_t *src = source->ptr;
+	ae_int16x4 gain_vec;
+	ae_int64 tmph, tmpl;
+
+	gain_vec = AE_L16_I((ae_int16 *)&gain, 0);
+
+	assert(mixed_samples >= start_sample);
+	samples_to_mix = AE_MIN_32_signed(mixed_samples - start_sample, sample_count);
+	samples_to_copy = sample_count - samples_to_mix;
+	n = 0;
+
+	for (left_samples = samples_to_mix; left_samples > 0; left_samples -= n) {
+		src = cir_buf_wrap(src + n, source->buf_start, source->buf_end);
+		dst = cir_buf_wrap(dst + n, sink->buf_start, sink->buf_end);
+		/* calculate the remaining samples*/
+		nmax = (int32_t *)source->buf_end - src;
+		n = AE_MIN_32_signed(left_samples, nmax);
+		nmax = (int32_t *)sink->buf_end - dst;
+		n = AE_MIN_32_signed(n, nmax);
+		in = (ae_int32x2 *)src;
+		out = (ae_int32x2 *)dst;
+		inu = AE_LA64_PP(in);
+		outu1 = AE_LA64_PP(out);
+		m = n >> 1;
+		left = n & 1;
+		/* process 2 samples per time */
+		for (i = 0; i < m; i++) {
+			AE_LA32X2_IP(in_sample, inu, in);
+
+			/* apply gain to in_sample */
+			in_sample32 = AE_SLAI32(in_sample, 8);	/* sign extension */
+			tmpl = AE_MUL32X16_L0(in_sample32, gain_vec);
+			tmph = AE_MUL32X16_H0(in_sample32, gain_vec);
+			tmpl = AE_SRAI64(tmpl, 8 + IPC4_MIXIN_GAIN_SHIFT);
+			tmph = AE_SRAI64(tmph, 8 + IPC4_MIXIN_GAIN_SHIFT);
+			in_sample = AE_SEL32_LL(AE_MOVINT32X2_FROMINT64(tmph),
+						AE_MOVINT32X2_FROMINT64(tmpl));
+
+			AE_LA32X2_IP(out_sample, outu1, out);
+			out--;
+			out_sample = AE_ADD24S(in_sample, out_sample);
+			AE_SA32X2_IP(out_sample, outu2, out);
+		}
+		AE_SA64POS_FP(outu2, out);
+
+		/* process the left sample to avoid memory access overrun */
+		if (left) {
+			AE_L32_IP(in_sample, (ae_int32 *)in, sizeof(ae_int32));
+
+			in_sample32 = AE_SLAI32(in_sample, 8);	/* sign extension */
+			tmpl = AE_MUL32X16_L0(in_sample32, gain_vec);
+			tmpl = AE_SRAI64(tmpl, 8 + IPC4_MIXIN_GAIN_SHIFT);
+			in_sample = AE_MOVINT32X2_FROMINT64(tmpl);
+
+			AE_L32_IP(out_sample, (ae_int32 *)out, 0);
+			out_sample = AE_ADD24S(in_sample, out_sample);
+			AE_S32_L_IP(out_sample, (ae_int32 *)out, sizeof(ae_int32));
+		}
+	}
+
+	for (left_samples = samples_to_copy; left_samples > 0; left_samples -= n) {
+		src = cir_buf_wrap(src + n, source->buf_start, source->buf_end);
+		dst = cir_buf_wrap(dst + n, sink->buf_start, sink->buf_end);
+		nmax = (int32_t *)source->buf_end - src;
+		n = AE_MIN_32_signed(left_samples, nmax);
+		nmax = (int32_t *)sink->buf_end - dst;
+		n = AE_MIN_32_signed(n, nmax);
+		in = (ae_int32x2 *)src;
+		out = (ae_int32x2 *)dst;
+		inu = AE_LA64_PP(in);
+		m = n >> 1;
+		left = n & 1;
+		for (i = 0; i < m; i++) {
+			AE_LA32X2_IP(in_sample, inu, in);
+
+			in_sample32 = AE_SLAI32(in_sample, 8);	/* sign extension */
+			tmpl = AE_MUL32X16_L0(in_sample32, gain_vec);
+			tmph = AE_MUL32X16_H0(in_sample32, gain_vec);
+			tmpl = AE_SRAI64(tmpl, 8 + IPC4_MIXIN_GAIN_SHIFT);
+			tmph = AE_SRAI64(tmph, 8 + IPC4_MIXIN_GAIN_SHIFT);
+			in_sample = AE_SEL32_LL(AE_MOVINT32X2_FROMINT64(tmph),
+						AE_MOVINT32X2_FROMINT64(tmpl));
+
+			AE_SA32X2_IP(in_sample, outu2, out);
+		}
+		AE_SA64POS_FP(outu2, out);
+		/* process the left sample to avoid memory access overrun */
+		if (left) {
+			AE_L32_IP(in_sample, (ae_int32 *)in, sizeof(ae_int32));
+
+			in_sample32 = AE_SLAI32(in_sample, 8);	/* sign extension */
+			tmpl = AE_MUL32X16_L0(in_sample32, gain_vec);
+			tmpl = AE_SRAI64(tmpl, 8 + IPC4_MIXIN_GAIN_SHIFT);
+			in_sample = AE_MOVINT32X2_FROMINT64(tmpl);
+
+			AE_S32_L_IP(in_sample, (ae_int32 *)out, sizeof(ae_int32));
+		}
+	}
+}
+
 #endif	/* CONFIG_FORMAT_S24LE */
 
 #if CONFIG_FORMAT_S32LE
@@ -264,17 +493,129 @@ static void mix_s32(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixe
 	}
 }
 
+static void mix_s32_gain(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixed_samples,
+			 const struct cir_buf_ptr *source,
+			 int32_t sample_count, uint16_t gain)
+{
+	int samples_to_mix, samples_to_copy, left_samples;
+	int n, nmax, i, m, left;
+	ae_int32x2 in_sample;
+	ae_int32x2 out_sample;
+	ae_int32x2 *in;
+	ae_int32x2 *out;
+	ae_valign inu = AE_ZALIGN64();
+	ae_valign outu1 = AE_ZALIGN64();
+	ae_valign outu2 = AE_ZALIGN64();
+	/* cir_buf_wrap() is required and is done below in a loop */
+	int32_t *dst = (int32_t *)sink->ptr + start_sample;
+	int32_t *src = source->ptr;
+	ae_int16x4 gain_vec;
+	ae_int64 tmpl, tmph;
+
+	gain_vec = AE_L16_I((ae_int16 *)&gain, 0);
+
+	assert(mixed_samples >= start_sample);
+	samples_to_mix = AE_MIN_32_signed(mixed_samples - start_sample, sample_count);
+	samples_to_copy = sample_count - samples_to_mix;
+	n = 0;
+
+	for (left_samples = samples_to_mix; left_samples > 0; left_samples -= n) {
+		src = cir_buf_wrap(src + n, source->buf_start, source->buf_end);
+		dst = cir_buf_wrap(dst + n, sink->buf_start, sink->buf_end);
+		/* calculate the remaining samples*/
+		nmax = (int32_t *)source->buf_end - src;
+		n = AE_MIN_32_signed(left_samples, nmax);
+		nmax = (int32_t *)sink->buf_end - dst;
+		n = AE_MIN_32_signed(n, nmax);
+		in = (ae_int32x2 *)src;
+		out = (ae_int32x2 *)dst;
+		inu = AE_LA64_PP(in);
+		outu1 = AE_LA64_PP(out);
+		m = n >> 1;
+		left = n & 1;
+		for (i = 0; i < m; i++) {
+			AE_LA32X2_IP(in_sample, inu, in);
+
+			/* apply gain to in_sample */
+			tmpl = AE_MUL32X16_L0(in_sample, gain_vec);
+			tmph = AE_MUL32X16_H0(in_sample, gain_vec);
+			tmpl = AE_SRAI64(tmpl, IPC4_MIXIN_GAIN_SHIFT);
+			tmph = AE_SRAI64(tmph, IPC4_MIXIN_GAIN_SHIFT);
+			in_sample = AE_SEL32_LL(AE_MOVINT32X2_FROMINT64(tmph),
+						AE_MOVINT32X2_FROMINT64(tmpl));
+
+			AE_LA32X2_IP(out_sample, outu1, out);
+			out--;
+			out_sample = AE_ADD32S(in_sample, out_sample);
+			AE_SA32X2_IP(out_sample, outu2, out);
+		}
+		AE_SA64POS_FP(outu2, out);
+
+		/* process the left sample to avoid memory access overrun */
+		if (left) {
+			AE_L32_IP(in_sample, (ae_int32 *)in, sizeof(ae_int32));
+
+			tmpl = AE_MUL32X16_L0(in_sample, gain_vec);
+			tmpl = AE_SRAI64(tmpl, IPC4_MIXIN_GAIN_SHIFT);
+			in_sample = AE_MOVINT32X2_FROMINT64(tmpl);
+
+			AE_L32_IP(out_sample, (ae_int32 *)out, 0);
+			out_sample = AE_ADD32S(in_sample, out_sample);
+			AE_S32_L_IP(out_sample, (ae_int32 *)out, sizeof(ae_int32));
+		}
+	}
+
+	for (left_samples = samples_to_copy; left_samples > 0; left_samples -= n) {
+		src = cir_buf_wrap(src + n, source->buf_start, source->buf_end);
+		dst = cir_buf_wrap(dst + n, sink->buf_start, sink->buf_end);
+		/* calculate the remaining samples*/
+		nmax = (int32_t *)source->buf_end - src;
+		n = AE_MIN_32_signed(left_samples, nmax);
+		nmax = (int32_t *)sink->buf_end - dst;
+		n = AE_MIN_32_signed(n, nmax);
+		in = (ae_int32x2 *)src;
+		out = (ae_int32x2 *)dst;
+		inu = AE_LA64_PP(in);
+		m = n >> 1;
+		left = n & 1;
+		for (i = 0; i < m; i++) {
+			AE_LA32X2_IP(in_sample, inu, in);
+
+			tmpl = AE_MUL32X16_L0(in_sample, gain_vec);
+			tmph = AE_MUL32X16_H0(in_sample, gain_vec);
+			tmpl = AE_SRAI64(tmpl, IPC4_MIXIN_GAIN_SHIFT);
+			tmph = AE_SRAI64(tmph, IPC4_MIXIN_GAIN_SHIFT);
+			in_sample = AE_SEL32_LL(AE_MOVINT32X2_FROMINT64(tmph),
+						AE_MOVINT32X2_FROMINT64(tmpl));
+
+			AE_SA32X2_IP(in_sample, outu2, out);
+		}
+		AE_SA64POS_FP(outu2, out);
+
+		/* process the left sample to avoid memory access overrun */
+		if (left) {
+			AE_L32_IP(in_sample, (ae_int32 *)in, sizeof(ae_int32));
+
+			tmpl = AE_MUL32X16_L0(in_sample, gain_vec);
+			tmpl = AE_SRAI64(tmpl, IPC4_MIXIN_GAIN_SHIFT);
+			in_sample = AE_MOVINT32X2_FROMINT64(tmpl);
+
+			AE_S32_L_IP(in_sample, (ae_int32 *)out, sizeof(ae_int32));
+		}
+	}
+}
+
 #endif	/* CONFIG_FORMAT_S32LE */
 
 const struct mix_func_map mix_func_map[] = {
 #if CONFIG_FORMAT_S16LE
-	{ SOF_IPC_FRAME_S16_LE, mix_s16 },
+	{ SOF_IPC_FRAME_S16_LE, mix_s16, mix_s16_gain },
 #endif
 #if CONFIG_FORMAT_S24LE
-	{ SOF_IPC_FRAME_S24_4LE, mix_s24 },
+	{ SOF_IPC_FRAME_S24_4LE, mix_s24, mix_s24_gain },
 #endif
 #if CONFIG_FORMAT_S32LE
-	{ SOF_IPC_FRAME_S32_LE, mix_s32 }
+	{ SOF_IPC_FRAME_S32_LE, mix_s32, mix_s32_gain }
 #endif
 };
 
diff --git a/src/audio/mixin_mixout/mixin_mixout_hifi5.c b/src/audio/mixin_mixout/mixin_mixout_hifi5.c
index 6d82ec54d50d..7c2942e76bd2 100644
--- a/src/audio/mixin_mixout/mixin_mixout_hifi5.c
+++ b/src/audio/mixin_mixout/mixin_mixout_hifi5.c
@@ -268,15 +268,16 @@ static void mix_s32(struct cir_buf_ptr *sink, int32_t start_sample, int32_t mixe
 
 #endif	/* CONFIG_FORMAT_S32LE */
 
+/* TODO: implement mixing functions with gain support!*/
 const struct mix_func_map mix_func_map[] = {
 #if CONFIG_FORMAT_S16LE
-	{ SOF_IPC_FRAME_S16_LE, mix_s16 },
+	{ SOF_IPC_FRAME_S16_LE, mix_s16, mix_s16 },
 #endif
 #if CONFIG_FORMAT_S24LE
-	{ SOF_IPC_FRAME_S24_4LE, mix_s24 },
+	{ SOF_IPC_FRAME_S24_4LE, mix_s24, mix_s24 },
 #endif
 #if CONFIG_FORMAT_S32LE
-	{ SOF_IPC_FRAME_S32_LE, mix_s32 }
+	{ SOF_IPC_FRAME_S32_LE, mix_s32, mix_s32 }
 #endif
 };